Virtual Screening Identifies Irreversible FMS-like ... - ACS Publications

Subscriber access provided by Macquarie University

Article

Virtual Screening Identifies Irreversible FMS-like Tyrosine Kinase 3 Inhibitors with Activity towards Resistance-conferring Mutations Dennis Bensinger, Daniel Stubba, Anjali Cremer, Vanessa Kohl, Theresa Waßmer, Johanna Stuckert, Victoria Engemann, Kimberly Stegmaier, Katja Schmitz, and Boris Schmidt J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01714 • Publication Date (Web): 11 Feb 2019 Downloaded from http://pubs.acs.org on February 12, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Virtual Screening Identifies Irreversible FMS-like Tyrosine Kinase 3 Inhibitors with Activity towards Resistance-conferring Mutations Dennis Bensinger†, Daniel Stubba†, Anjali Cremer‡, Vanessa Kohl†, Theresa Waßmer†, Johanna Stuckert†, Victoria Engemann†, Kimberly Stegmaier‡, Katja Schmitz† and Boris Schmidt†* †Clemens-Schöpf-Institute

for Organic Chemistry and Biochemistry, Technische Universität

Darmstadt, 64287 Darmstadt, Germany ‡Department

of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston,

Massachusetts 02215, United States

Abstract The use of covalent irreversible binding inhibitors is an established concept for drug development. Usually the discovery of new irreversible kinase inhibitors occurs serendipitously showing that efficient rational approaches for the rapid discovery of new drugs are needed. Herein, we report a virtual screening strategy that led to the discovery of irreversible inhibitors of the FMS-like tyrosine kinase 3 (FLT3) involved in the pathogenesis of acute myeloid leukemia (AML). A virtual screening library was designed to target the highly conserved Cys828 residue preceding the DFG motif by modification of reported reversible inhibitors with chemically reactive groups. Prospective covalent docking allowed

ACS Paragon Plus Environment

1

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 85

the identification of two lead series, resulting in a massive increase in inhibition of kinase activity and cell viability by irreversible inhibitors compared to the corresponding reversible scaffolds. Lead compound 4b (BSc5371) displays superior cytotoxicity in FLT3-dependent cell lines to compounds in recent clinical trials and overcomes drug-resistant mutations.

Introduction Acute myeloid leukemia (AML) is a malignant neoplasia affecting hematopoietic progenitor cells. These cells fail to differentiate into functional blood cells of the myeloid path (erythrocytes, neutrophils and thrombocytes), while ongoing cell proliferation leads to the accumulation of myeloblasts in the bone marrow.1 The most common pathogenic mutations occur on the gene encoding for FMS-like tyrosine kinase 3 (FLT3). Length mutations in this gene through internal tandem duplications (ITD) in the juxtamembrane domain occur in about 25% of patients and lead to ligand-independent auto-phosphorylation and downstream signaling.2 Point mutations in the FLT3 kinase domain (e.g. D835Y) occur in 10% of patients and destabilize the inactive kinase conformation leading to higher kinase activity.3 In 2017 the multitargeted kinase inhibitor Midostaurin (PKC412) was approved by the US Food and Drug Administration (FDA) in combination with chemotherapy for the treatment of de novo FLT3-mutated AML showing a 22% reduction of risk of death.4 Minimal residual disease after treatment with tyrosine kinase inhibitors (TKIs) like sorafenib and quizartinib can lead to relapse and is often associated with resistance-conferring point


2

Page 3 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


mutations in an FLT3-ITD allele.2 As emergence of drug-resistant mutations is a common liability of targeted therapies, new agents with distinct properties are needed. The use of irreversibly binding TKIs has several advantages over classical reversible inhibitors such as an increased target residence time, altered selectivity, as well as more durable responses than their reversibly binding counterparts while idiosyncratic toxicity and sustained off-target effects are additional risks.5 They act by the formation of a covalent bond between an amino acid with a nucleophilic side chain such as cysteine and a chemically reactive electrophilic group (CRG) connected to a specific ligand. Bond formation usually takes place after a strong non-covalent complex has formed that positions the CRG in close vicinity to the target nucleophile. Upon covalent bond formation, irreversible reactions rule out ligand dissociation leading to non-equilibrium inhibition kinetics. Thus, the enzyme function can be restored by de novo synthesis only. So long, four TKIs are approved by the FDA for the treatment of certain malignancies, afatinib,6 neratinib,7 osimertinib,8 and ibrutinib,9 act by irreversibly inhibiting the EGFreceptor family and Bruton´s tyrosine kinase, respectively. The natural product hypothemycin is a resorcylic acid lactone that irreversibly inhibits 18 kinases, including FLT3.10 The cis-enone motif is attacked by a cysteine directly preceding the DFG motif at the start of the activation loop (Cys828 in FLT3). These so-called “group 4” cysteines are the most conserved in the human kinome occurring in 48 of about 500 kinases.11 A crystal structure analysis of hypothemycin and ERK2 confirms covalent binding


3


Page 4 of 85

in the DFG-in conformation.12 Despite the advantages of irreversible inhibitors, only few drug development efforts have focused on targeting this group of cysteines. Examples include benzoquinone-substituted quinazolines targeting VEGFR213 and analogues or ring-opened derivatives of resorcylic acid lactones targeting FLT3, showing only micromolar cellular activity.14 Recently, discovery of covalent binding inhibitors was advanced by the development of molecular modelling algorithms that allow covalent docking.15 In 2015, our group introduced the covalent docking tool DOCKTITE16 as a freely available script package for the Molecular Operating Environment.17 In this method a CRG is tagged by a pre- or user-defined transformation, that is then attached to a selected covalent attachment point of the receptor. During docking the nucleophilic residue is detached from the receptor but fixed by a tight pharmacophore. This strategy allows highly customizable virtual screening strategies while all docking steps can be monitored and controlled. Consequently, DOCKTITE has been successfully used for the development of covalent inhibitors for the cysteine proteases rhodesain,18-19 cruzain18 and falcipain20 as well as for the plasmodium proteasome.21 Despite the recent progress described, only one case study identifying covalent kinase inhibitors in silico has been reported so far. London et al. successfully used the DOCKovalent tool to identify covalent reversible inhibitors of the kinases RSK2 and JAK322 as well as irreversible MKK7 inhibitors, that were published while this study was under revision.23 However, the scope of covalent docking methodology has to be enlarged, as only cyano acrylamides and


4

Page 5 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


acrylamides were evaluated and hit identification profited by available protein-ligand structures with the modified target cysteine (RSK2) or numerous cocrystal structures with related hinge binders (JAK3, MKK7), respectively. Because of the impressive clinical benefits of irreversible inhibitors and the vulnerability of FLT3 treatment to resistance we sought to rationally design and evaluate irreversible FLT3 inhibitors. In the following, we present a strategy that allows the identification of irreversible FLT3 inhibitors with superior cytotoxicity to inhibitors in clinical trials that can overcome resistance-conferring mutations of FLT3.

Results and Discussion Rational Design of Irreversible FLT3 Inhibitors. A major drawback of FLT3 inhibitors in recent clinical trials was the acquisition of point mutations within the ATP-binding site thus we decided to target FLT3 in its catalytically active DFG-in conformation. Within the ATPbinding site FLT3 possesses three targetable cysteine residues. Two of them, Cys694 (gatekeeper + 3) and Cys695 (gatekeeper + 4), are part of the hinge region while Cys828 is preceding the conserved DFG-motif. Kinase inhibitors usually gain potency by forming alternating hydrogen bond donor/acceptor interactions with the hinge backbone atoms. Therefore modification of known hinge binding motifs are promising starting points to target the neighboring Cys828 residue. Additionally, exposure of Cys828 to the surface of the ATP


5


Page 6 of 85

binding pocket allows variation of the trajectory by which a CRG is recognized and a covalent bond can be formed. However, no crystal structure of FLT3 in the active conformation has been reported, so we designed a homology model of the active FLT3 conformation using c-KIT (pdb 1PKG) as a template (Figure S1-S3).24

Figure 1. (a) Virtual screening and filter strategy for the in silico discovery of irreversible FLT3 inhibitors; (b, c) Structure of CRGs and pose distribution among scaffolds. HA: α-halo acetamide, AC: acrylamide, VS: vinyl sulfonamide and PA: propargyl amide (d) Rational design and docking hits of the bisaminopyrimidine scaffold 3 and (e) the indolinone inhibitor scaffold 6 (gk = gatekeeper residue).


6

Page 7 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


c-KIT is the most closely related kinase to FLT3 with a sequence similarity of 79% and a sequence identity of 64% and consequently numerous shared potent inhibitors.25 To validate the homology model, we docked hypothemycin covalently to Cys828 of FLT3. Rigid body alignment of the top-scored conformation docked into FLT3 with the conformation obtained from the co-crystallized structure in ERK2 gave a r.m.s. deviation (RMSD) of 1.34 Å, while the top-scored redocked pose in ERK2 showed a RMSD of 1.08 Å (Figure S4). To identify potent reversible inhibitors as starting points for reactive modifications and subsequent covalent docking, the publicly accessible BindingDB Database was manually searched for FLT3 inhibitors with distinct hinge binding motifs (Figure 1a, 1752 entries as of Jan. 2015).26 Typical inhibitors that were excluded contain bisarylurea motifs such as sorafenib, quizartinib and ponatinib that are designed to occupy the hydrophobic pocket II and thus bind to the inactive DFG-out conformation.2 We found a set of 13 different hinge binding scaffolds (Figure S5) and identified a plausible reversible binding pose for each scaffold (Figure S6 – S18). Those scaffolds form mono- (5 scaffolds), bi- (6 scaffolds) or tridentate (2 scaffolds) alternating hydrogen bond donor/acceptor interactions to the gatekeeper+1 (Glu692) and +3 residue (Cys694) that are comparable to hinge binding observed from crystal structures of related scaffolds.27 We selected four chemically and sterically distinct CRGs (Figure 1b) to be attached to the non-covalent docked structure. The acrylamide (AC), vinyl sulfonamide (VS) and propargyl amide (PA) groups span approximately 5 Å, while the α-halo acetamide (HA) warhead spans only 3.8 Å (Figure S19). We concluded, that CRGsubstitution of aromatic ring carbons within 5 Å to the Cys828 sulfur may enable covalent


7


Page 8 of 85

binding (Figure S6F to S18F), while favorable conformations were selected by covalent docking. Subsequently, we identified 32 attachment points among the selected scaffolds leading to a targeted virtual library of 128 compounds (Table S1). Since addition of nucleophiles to the propargyl amide can lead to E- or Z- substituted alkenes, addition to the covalent attachment point prior to pharmacophore docking gave 160 docking input structures. Virtual Screening and Filter Strategy. Covalent docking was performed using the pharmacophore placement method. The pharmacophore features of the 13 scaffolds were defined by the aryl, donor and acceptor groups forming hinge interactions and complemented with a tight pharmacophore at the target cysteine (Figure S6E to S18E). However, the pharmacophore feature of the nucleophilic sulfur atom in Cys828 was less stringent (d = 3.0 Å) allowing different rotamers. The initial docking gave 831 poses (Table S2, Figure S20) that were filtered for Glu692 and Cys694 backbone hydrogen bond donor acceptor interactions (404 poses left) and subsequently rescored using the DSX scoring function28 while only the top 5 poses of every ligand were retained (319 poses left). We then calculated the ligand efficiency based on the London dG score and set a cutoff at -0.5 kcal/mol/heavy atom to limit the search to highly efficient inhibitors (102 poses left, Figure 1c, Table S3 – S5). From the reversible scaffolds, 5 scaffolds retain more than 10 hits of the corresponding reactive inhibitors (73% of total hits). On average, 32% of the initial docking hits in this group ranging from 13% (bisaminopyrimidine, scaffold 3) to 67% (indolinone,


8

Page 9 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


scaffold 6) are retained by the filter strategy. The hit distribution varies widely among the CRGs. With 45% of total hits vinyl sulfonamides are the most abundant group, followed by the α-halo acetamides (29%), acrylamides (16%) and propargyl amides (10%). Characterization of Reversible Binding Fragments. As the potency of irreversible binding inhibitors is based on a moderate affinity between target and ligand in a reversible equilibrium prior to inactivation, we sought to evaluate the potency and structure-activity relationship of the selected reversible hits to further increase efficiency and the success rate of prospective irreversible inhibitor design. As the rational design starts from reported reversible ligands, synthetic procedures used to prepare reversible compounds 1a to 1g were based on earlier work (see experimental section). Among the initial series of reversible inhibitors (1a to 1e), bisaminopyrimidine 1a is the most cytotoxic compound (89% cell death, 10 µM, Table 1) and strongest inhibitor of FLT3wt. The indolinone 1b shows a comparable inhibition for FLT3wt and FLT3(D835Y). Imidazopyridine 1c and quinazoline 1d are less potent inhibitors of FLT3(D835Y) while azaindole 1e shows only weak inhibition of FLT3wt. However, all of the compounds of the initial series show only low cytotoxic activity in FLT3ITD mutant MV4-11 cells. To increase reversible binding affinity, we applied small changes to the most promising inhibitors 1a and 1b that are compatible with the initial, reversible docking poses leaving the attachment points for the CRGs unchanged (Figure 1d, e; S21a, b). Exchange of the phenyl ring in 1b with a 2-pyrrole (1g) constrains a planar conformation of the ligand by formation of an intramolecular hydrogen bond. The 5-methyl group connected


9


Page 10 of 85

to the pyrazolyl amine 1a was changed to a cyclopropyl ring (1f) that has also been reported to efficiently target FLT3.29 These small modifications massively increase the cytotoxicity of both scaffolds as well as inhibition of FLT3 and FLT3(D835Y) kinase activity making these fragments an efficient starting point for the development of irreversible inhibitors. The same modifications were applied to the poses obtained from the initial covalent docking and subsequent energy-minimization leaves the poses largely unchanged (Figure S21c – j).


10

Page 11 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1. Inhibition of FLT3wt; FLT3(D835Y) and MV4-11 cell viability treated with reversible binding fragments. a Eurofins KinaseProfiler, b ADP-Glo Assay. IC50 Structure

Compound

(MV4-11) / nM

H N N

N

N

FLT3wt a

FLT3 (D835Y)b

NH

HN

H N

O

Relative Inhibition (cTKI = 1 µM) / %

1a

> 1.000 nM

82

15

1b

> 1.000 nM

60

45

1c

> 1.000 nM

76

8

1d

> 1.000 nM

26

25

1e

> 1.000 nM

25

83

1f

230±53

98

88

1g

380±21

97

86

N N

N

O

N

O HN

H N

N

H N N

N

N

HN

NH

O

H N

NH


11


Page 12 of 85

Characterization of Indolinone-based Irreversible Inhibitors. A total of nine inhibitors based on the reversible indolinone hit 1g were synthesized (Table 2). In addition to the CRGsubstituted derivatives we synthesized the analogous unreactive inhibitors to assess the effect of the amide or sulfonamide linking group on reversible binding affinity. To evaluate the hit rate of the screening strategy, we also synthesized the inhibitors that were not identified as a hit during virtual screening. The 5-substituted vinyl sulfonamide 2i was the most potent inhibitor of this series showing a 5-fold in-crease in inhibition of MV4-11 cell viability compared to the initial fragment as well as the highest inhibition of kinase activity in FLT3(D835Y) and FLT3(ITD) in this series. Importantly, the unreactive ethyl sulfonamide 2h shows a massive decrease in inhibition of cell viability and kinase activity with decreased inhibition in FLT3(D835Y) and FLT3(ITD) suggesting that potency arises from covalent bond formation as both compounds are able to adopt the same reversible conformations. The vinyl sulfonamide warhead gave the highest number of hits in silico with the top virtual ligandefficiency (LEV) at rank 6 of the 102 hits (LEV = -0.55) and further poses on ranks 16, 22, 41 and 67. The propargyl amide 2g as well as acrylamide 2f shows a massive loss of cytotoxic activity compared to the initial hit similar to the unreactive propione amide 2e and low inhibition in kinase assays.


12

Page 13 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 2. Inhibition of cell viability of MV4-11 cells and FLT3-mutant kinase activity by the indolinone fragment 1g and the corresponding 5-substituted indolinones 2a – i with chemically reactive electrophilic groups and the analogous unreactive groups.

Relative Inhibition / % IC50 (MV4-11) / nM

c = 0.1 µM

c = 1 µM

c = 0.1 µM

1g

380 ± 21

40

86

61

2a

> 1.000

47

90

33

F

2b

845 ± 20

57

92

51

Cl

2c

72 ± 8

53

87

61

Br

2d

653 ± 200

48

92

21

2e

> 1.000

15

81

44

2f

> 1.000

29

79

25

2g

709 ± 50

5

58

55

O S N H O

2h

> 1.000

13

72

9

O S N H O

2i

67 ± 8

77

96

93

N H O

R N H

Compound

R= -H

FLT3(D835Y) FLT3(ITD)

O N H O N H

O N H O N H

O N H O N H O N H


13


Page 14 of 85

Thus, covalent inhibition is not favorable for acrylamide 2f though it had been identified in a single pose during virtual screening (rank 1, LEV =

-0.61) obtaining a similar

conformation as 2i. The α-halo acetamides show comparable inhibition of kinase activity in FLT3(D835Y) with the unsubstituted acetamide while a slight preference for the α-chloro acetamide was found inhibiting FLT3(ITD). While the reversible acetamide-substituted indolinone fragment 2a shows no inhibition of MV4-11 cell viability, the α-fluoro- (2b) and α-bromo-substituted (2d) acetamides are less active than the initial screening hit 1g. The αchloro acetamide 2c shows similar cytotoxicity as the vinyl sulfonamide 2i. Having established the structure-activity relationship of indolinone-based irreversible inhibitors we reanalyzed the results from covalent docking to explain disagreements between predicted and confirmed hits. A possible cause for the false-positive hit acrylamide 2f might be the negligence of the covalent bond formation as only the reversible fragment and the final covalent complex are modeled during covalent docking. The energy profile of the dihedral angle between the C-CN plane of the aromatic amine and the C-N-S plane of the reactive group in the top-scored pose of the active vinyl sulfonamide 2i shows that the energy is close to a local minimum while the dihedral angle is only changed by 3° (56° to 59°) after energy-minimization of the unbound ligand (Figure 2a, S22). The position of the targeted distal carbon atom is slightly shifted by 0.72 Å giving an overall RMSD of 0.49 Å. In contrast, the position of the distal carbon atom in the top-scored docking pose of the inactive acrylamide 2f is changed by 2.24


14

Page 15 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Å while the dihedral angle is changed by 39° (overall RMSD 0.89 Å) explaining that the identified hit could not be confirmed experimentally as the modeled conformation of the CRG is unfavorable (Figure 2b). Furthermore, conformational search of unbound ligands 2f and 2i shows that the acrylamide warhead is oriented in-plane of the indolinone ring, while the vinyl sulfonamide retains multiple conformations out-of-plane enabling induced fit to the ATP binding site and vicinity to Cys828 (Figure S23). We further assume that the initial pharmacophore distance of 1.5 Å might be too stringent to allow binding of all warheads, as the α-halo acetamide series was a false negative hit. Consequently, redocking of the α-halo acetamide derivatives 2b to 2d was performed using 2 Å ligand pharmacophores. This adjustment led to numerous plausible docking poses and thus showed to be predictive for covalent inhibition by α-halo acetamides (Figure S24).


15


Page 16 of 85

Characterization of Bisaminopyrimidine-based Irreversible Inhibitors. Similar to the indolinone scaffold, we synthesized 14 inhibitors of the 3- and 4-substituted bisaminopyrimidine scaffold including derivatives with all four types of CRG to include structures predicted as hits and predicted negatives as well as the corresponding reversible substitutions (Table 3). Similarly, the vinyl sulfonamide derivatives 3m and 3n of the initial fragment 1f show massively increased inhibition of MV4-11 cell viability while the corresponding reversible structures 3k and 3l show similar cytotoxicity as the reversible fragment 1f. Inhibition of FLT3 activity shows a comparable increase in inhibition for the reactive derivatives towards FLT3(D835Y) and FLT3(ITD). Interestingly, 3m shows lower inhibition of FLT3(ITD) but 25-fold increased cytotoxicity in MV4-11 cells suggesting synergistic off-target effects. The 3-methyl derivatives corresponding to 3m (2 hits, rank 51, 96) and 3n (3 hits, rank 17, 54, 83) have been identified during virtual screening and redocking of 3m and 3n gave similar conformations (Figure S21k – n). These results show that the screening protocol chosen allows predictive hit finding for the bisaminopyrimidine scaffold as well.


16

Page 17 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 3. Inhibition of cell viability of MV4-11 cells and FLT3-mutant kinase activity by the bisaminopyrimidine fragment 1f and the corresponding 3- and 4-substituted inhibitors 3a – n with chemically reactive electrophilic groups or the analogous unreactive groups.

H N N

N

N

Relative Inhibition / % NH

Compound

HN

Substitution

R

IC50 (MV4-11) / nM

FLT3(D835Y)

FLT3(ITD)

c = 0.1 µM

c = 1 µM

c = 0.1 µM

R= -H

1f

-

230 ± 53

52

88

27

3a

3

> 1.000

36

85

54

3b

4

> 1.000

45

86

82

3c

3

175 ± 2

33

90

74

3d

4

141 ± 13

39

93

74

3e

3

> 1.000

37

90

66

3f

4

> 1.000

34

80

22

3g

3

787 ± 100

22

81

50

3h

4

521 ± 65

29

84

59

3i

3

> 1.000

38

58

2

N H

3j

4

> 1.000

45

54

17

O S N H O

3k

3

489 ± 17

32

92

59

3l

4

129 ± 16

72

96

86

O S N O H

3m

3

2.3 ± 0.1

47

90

73

3n

4

58 ± 16

86

95

99

O N H

O Cl

N H O N H

O N H O


17


Page 18 of 85

The acrylamide substitution shows no benefit in cytotoxicity compared to the reversible fragment 1f while the corresponding reversible inhibitors 3e and 3f are almost inactive. During virtual screening the acrylamides have been identified as low-ranked compounds only (3g: rank 99 of 102, 3h: rank 49, 101). The 3- and 4- substituted propargylamides 3i and 3j are the least active inhibitors of this series displaying only neglectable MV4-11 cytotoxicity. Although the propargyl amide 3j yielded no poses during screening, 3i was identified (rank 10, 45). Reversible acetamide-substitution of the initial screening hit leads to a loss of cytotoxicity in 3- (3a) as well as 4-position (3b) while the α-chloro substitutions in 3c and 3d leads to a slightly increased cytotoxicity compared to fragment 1f suggesting a covalent mode of action. However, inhibition of FLT3(D835Y) and FLT3(ITD) was comparable to reversible acetamides 3a and 3b, a trend already observed with the indolinone α-chloro-acetamide 2c, suggesting that kinase screening alone would under-estimate the overall potential of this CRG. In summary, the screening strategy presented gave a hit rate of 73% based on the number of predicted poses (16 correct-positive, 6 false-positive, Figure S25) where all of the vinyl sulfonamide derivatives (3 hits, 10 poses) are validated. Design of 2nd Generation Indolinone Inhibitors. As the design of covalent binding inhibitors was based on fragmentation of larger lead compounds, we aimed to increase FLT3 inhibition and cytotoxicity by merging the most active irreversible fragments with the parent scaffold. The bisaminopyrimidine scaffold is less suitable for this purpose as the pyrazolyl amine motif may lead to promiscuous kinase inhibitors.30


18

Page 19 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Exemplary, we tested activity of the reversible inhibitors 1f and 1g on the far-distant kinases GSK3α and GSK3β yet containing the DFG-1 Cys residue, showing potent inhibition for 1f (74% and 76% at 1 µM) but only insignificant inhibition by 1g (12% and -6% at 1 µM). Furthermore, inhibition of cell viability is about 10-fold higher for the meta substituted vinyl sulfonamide 3m than for the isomer 3n while inhibition of FLT3 kinase activity shows the opposite trend suggesting that cytotoxicity may arise from synergistic effects by off-target inhibition. Furthermore, 3m and 3n are potently cytotoxic towards FLT3wt THP1 cells and FLT3-negative Jurkat cells (Table 4). Additionally, ligand efficiency in MV4-11 cells is higher for the indolinone derivatives 2c (-0.34) and 2i (-0.33) than for the bisaminopyrimidine derivatives 3m (-0.29) and 3n (-0.26). Furthermore, the top-scored pose of redocked fragments 2c and 2i resemble the orientation of sunitinib in cocrystal structures, where the 3-carbon of pyrrole is exposed to the solvent site of the ATP binding pocket (Figure S26). Irreversible indolinone scaffolds 2c and 2i thus were merged with the parent inhibitor sunitinib by addition of a 2- and 4-methyl group to the pyrrole ring as well as a 3carboxamide-linked ethylene diethylamine group to give the hybrid structures 4a and 4b (Figure 2c). Synthesis was performed similar to procedures described before (SI chapter 2h). Redocking in FLT3 DFG-in was performed using a 2 Å pharmacophore of the indolinone hydrogen bond donor, acceptor and aryl feature. The top-scored binding poses of 4a and 4b show comparable binding modes as fragments 2c and 2i and sunitinib (Figure 2d, e; S27, S28). The conformation of the solvent-exposed diethylamine group is highly flexible (Figure S29). Receptor-free energy minimization of the ligand shows a distance of the vinyl Cβ atom of


19


Page 20 of 85

1.12 Å to its position in the docked pose while the energy plot as a function of the dihedral angle of the sulfonamide is close to its minimum (Figure S30).

Figure 2. Comparison of the docked (green) and energy-minimized structure (blue) of (a) the active vinyl sulfonamide 2i and (b) the inactive acrylamide 2f. (c) Structure of 2ndgeneration indolinone inhibitors. (d) Top-scored covalent docking pose of 4a and (e) 4b.


20

Page 21 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 4. Inhibition of cell viability in FLT3-mutated (MV4-11), FLT3wt (THP1) and FLT3negative (Jurkat) cells after 72 h treatment. Inhibition of purified kinase activity is determined at 100 µM ATP and 30 min pre-incubation of kinase and inhibitor using the ADP Glo assay. a Half-life towards 100-fold excess of reduced glutathione and 10 µM inhibitor as determined by HPLC. n.d. = not determined.

Compound

MV4-11 (FLT3(ITD))

IC50 / nM THP1 (FLT3wt)

Jurkat (FLT3 neg.)

Relative Inhibition / % FLT3(D835Y) FLT3(ITD) c = 0.1 µM c = 1 µM c = 0.1 µM

4a

21 ± 1

> 1 µM

> 1 µM

71

90

73

218

4b

6.0 ± 0.3

> 1 µM

> 1 µM

95

100

86

51

4c

> 1 µM

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

4d

643 ± 95

n.d.

n.d.

32

n.d.

n.d.

n.d.

3m

2.3 ± 0.1

172 ±24

44 ± 5

47

90

73

58

3n

58 ± 16

514 ± 8

298 ± 10

86

95

99

82

Sunitinib

54 ± 1

> 1 µM

> 1 µM

82

96

70

n.d.

Hypothemycin

15 ± 3

> 1 µM

> 1 µM

85

92

91

n.d.

Crenolanib

12 ± 3

> 1 µM

> 1 µM

95

100

97

n.d.

t1/2 / mina

Introduction of the 4-methyl group close to Cys694 changed the position of the sulfonamide group, slightly forcing a strong hydrogen bond donor-acceptor interaction with Lys644. Molecular dynamics simulation of 4b in complex with the FLT3-DFG-in homology model over 110 ps gives only small RMS deviations (0.5 Å to 1.0 Å) compared to the initial complex concluding that the protein-ligand complex is highly stable (Figure S31). The vinyl sulfonamide substituted sunitinib derivative 4b shows an increase in cytotoxicity towards


21


Page 22 of 85

MV4-11 cells compared to the irreversible fragment 2i while the reversible ethyl sulfonamide 4d is ~100-fold less cytotoxic. Additionally, inhibition of FLT3(D835Y) is further increased, while inhibition decreases by the ethyl sulfonamide compared to sunitinib. Optimization of the irreversible fragment 2c to give the sunitinib-derived α-chloro acetamide 4a yields a 2.5-fold increased cytotoxicity while the unreactive acetamide derivative 4c is inactive in MV4-11 cells. 4a and 4b exhibit both superior cytotoxicity to sunitinib in MV4-11 cells while they retain selectivity towards FLT3wt THP1, NOMO1, U937 as well as FLT3-negative Jurkat cells (Figure S32). In FLT3(ITD)-positive MOLM-14 cells 4b (IC50 = 7.8 nM) is more cytotoxic than 4a (IC50 = 26.4 nM) and 3m (IC50 = 11.6 nM). Vinyl sulfonamide 4b shows superior cytotoxicity compared to the type I inhibitors midostaurin (~2x), crenolanib (~2x) and hypothemycin (~2.5x) in MV4-11 cells. While increased cytotoxicity in MV4-11 cells by 4b compared to 4a correlates with decreased pFLT3 and pSTAT5 levels, 3m shows higher levels of prosurvival-signaling correlating with decreased inhibition in FLT3 kinase assays (Figures 3a, S33).

Reactivity towards Nucleophiles. Half-life of electrophilic compounds was compared in a HPLC-based assay using 100-fold excess of nucleophiles to investigate the properties of the different CRGs (Figures 3b). While the acrylamide 2f and α-fluoroacetamide 2b are highly stable (t1/2 > 24 h), the propargylamide 2g (50 min) as well as the hit-compounds 2c (384 min) and 2i (71 min) are less stable. Half-life is decreased by the introduction of the alkaline tertiary amine in 4a (218 min) and 4b (52 min). Consequently, pre-incubation of 4b with a


22

Page 23 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1000x excess of glutathione (100 nM inhibitor) for 30 min decreased the inhibition of FLT3(D835Y) kinase activity from initial 95% to 68% while incubation for 24h decreases inhibition further (37%). Increase of buffer pH leads to a decrease of half-life (28 min, pH 8) while a decrease of pH increases half-life (130 min, pH 6.8; Figure S34). The α-N-Cbzprotected lysine, serine and threonine amino acids show no conversion with 4b (pH 7.4).

Danio Rerio Embryo Toxicity Assay. Biocompatibility and toxicologic effects of irreversible FLT3 inhibitors were investigated in Danio rerio embryos, that have been shown to be a promising model for drug toxicity screening (Figure 3c).31 Interestingly, in Danio rerio the targeted cysteine (DFG-1) is replaced by an alanine residue, suggesting that off-target rather than FLT3-dependent toxicity can be observed (Table S6, S7). Comparison of the 2-halosubstituted acetamides shows high toxicity for the bromo- 2d and fluoro- 2b derivative (10 µM inhibitor), while the chloro- derivative 2c as well as the reversible acetamide 2a are not lethal up to 96 h post incubation (120 hpf). The vinyl sulfonamide 2i as well as hybrid structures 4a and 4b are non-toxic at 10 µM comparable to sunitinib, hypothemycin and crenolanib. The bisamino-pyrimidine 3m shows increased toxicity that is comparable to fragment 1f suggesting a more promiscuous mode of action.


23


Page 24 of 85

Figure 3. (a) Western Blot analysis of FLT3(Y969) phosphorylation and downstream STAT5 phosphorylation after 1 h incubation with 4b. (b) Half-life of electrophilic inhibitors towards reduced glutathione (100x excess, pH 7.4). (c) Toxicity of selected inhibitors on Danio rerio embryos (96 hpf, 72 h incubation, 10 µM). (d) Inhibition of FLT3(D835Y) by 4a and 4b is insensitive to high ATP concentrations but is (e) highly dependent on ATP concentration for reversible inhibitors.

ATP- and time-dependent inhibition. We further investigated the ATP- and timedependency of kinase inhibition by compounds 4a, b and 3m, n in FLT3(D835Y) representing the most active compounds from both series. As expected, the reversible binding FLT3 inhibitors crenolanib and sunitinib show a strong dependency on ATP concentration, where a 10-fold ATP increase (100 µM to 1 mM) leads to an approximately 10-fold decrease of inhibition whereas crenolanib (4.2 nM to 42 nM) remains more potent


24

Page 25 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


than sunitinib (8.2 nM to 105 nM, Figure 3d, e). Since irreversible bond formation leads to non-equilibrium-binding only the initial reversible complex, preceding covalent bond formation, is ATP-competitive. Indeed, inhibition of FLT3(D835Y) by 4a (67 nM at low ATP, 58 nM at high ATP) as well as 4b (11.6 nM at low ATP to 18.6 nM at high ATP) is less vulnerable to increased ATP concentrations. To our surprise, 3m and 3n are highly susceptible to high ATP concentrations suggesting low inactivation kinetics and thus high competitivity to ATP (Figure S35). Additionally, we determined the influence of the preincubation time of inhibitor and kinase alone where ATP is not competing with inhibitor binding. As expected, the indolinones 4a and 4b as well as the bisaminopyrimidines 3m and 3n show increased inhibition of FLT3(D835Y) and FLT3(ITD) when pre-incubation time is increased from 5 min to 30 min (Table S8). To determine the Ki and kinact values of the most potent covalent FLT3 inhibitors 4a and 4b we evaluated time-dependent IC50 values against FLT3(D835Y) (Figure S36). Ki and kinact have been obtained from the plot IC50(t) vs t by nonlinear regression (see experimental section). We found that 4a has a lower binding affinity (Ki = 10.14 nM) than 4b (Ki = 1.98 nM) but a higher inactivation constant (kinact = 0.0033 / s-1 and 0.0017 / s-1). However, determination of Kinact/Ki shows that 4b is a more potent irreversible inhibitor (kinact/Ki = 0.84 µM-1 s-1) than 4a (kinact/Ki = 0.33 µM-1 s-1).”

Trypsin-digest and MALDI-MS of covalently modified FLT3. Since compounds 4a and 4b represent the most active compounds in this study covalent modification of Cys828 was confirmed by tryptic in-gel digest of FLT3wt, that was considered to yield the modified


25


Page 26 of 85

octapeptide IC828DFGLAR that could be detected by MALDI mass spectrometry. Standard denaturation and alkylation protocol using dithiothreitol and 2-iodo acetamide allowed detection of the single acetamide-substituted peptide with good agreement of the calculated and found masses and the modeled and measured isotopic pattern (Figure S37 – S41). Subsequent MS/MS allowed detection of numerous fragments of this sequence. Preincubation of FLT3wt with 2i, hypothemycin, 4a and 4b allowed detection of calculated masses with small errors confirming covalent modification of Cys828 by the key compounds in this study.

Kinase selectivity of 4b and activity towards resistance-conferring mutations. We chose compound 4b as lead structure in this project as it shows the strongest inhibition of the isolated kinase and strongest cytotoxicity and further investigated kinase selectivity as well as activity in drug-resistant mutations. Selectivity of the most potent inhibitor 4b was determined in a panel of 34 kinases at 100 nM inhibitor (Figure 4a). Kinases, that are inhibited by sunitinib, were selected for testing based on available Kd values for sunitinib and for containing the homologous CDFG motif (Table S9, Figure S42). As expected, FLT3 (95%) and FLT3(D835Y) (99%) are potently inhibited. Surprisingly, other kinases of the PDGF receptor family containing the conserved DFG-1 cysteine are significantly less inhibited, including c-KIT (34%), PDGFRα (24%) and PDGFRβ (14%). This is remarkable as they all share high sequence similarities (>70%) to FLT3 and exhibit single-digit to sub-nanomolar Kd values for sunitinib.32 Interestingly, less related kinases (30 to 40% sequence similarity)


26

Page 27 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


containing the CDFG motif (ERK1, ERK2, GSK3α, GSK3β) are not inhibited at 100 nM 4b while being reported targets of hypothemycin.10 4b may also covalently modify the homologous serine residue (DFG-1) as the Ret kinase (96%) is strongly inhibited. Interestingly, 5-amino indolinones have been reported to be potent Ret and FLT3 inhibitors while superposition of the FLT3 model and Ret cocrystal structure is in good accordance (RMSD = 0.7 Å, Figure S43).33 The homolog alanine and glycine residues are tolerated by 4b (Lyn, Lck, Axl, IRAK1; 93% to 81% inhibition) suggesting that hydrogen bond interactions to the conserved lysine may account for increased affinity compared to sunitinib while a steric clash with CysSγ or SerOγ is prevented. We further assessed our lead compound 4b using the KinomeScan screening platform, where numerous FLT3 inhibitors have been assessed before (Figure 4b).34-35

Figure 4. (a) Selectivity of 4b in a panel of 34 kinases.36 (b) Kd values of 4b and inhibitors in clinical trials in a panel of FLT3-mutants. (c) Relative inhibition by time (IC50(crenolanib)/


27


Page 28 of 85

IC50(4b)). (d) Effect of inhibitor wash-out after 2 h, 8 h, 24 h incubation and read-out at 72 h in MV4-11 cells.

Sunitinib and crenolanib show higher affinity towards FLT3wt and FLT3(ITD) than 4b, while midostaurin is less efficient. Interestingly, sunitinib showed a ~6-fold increased Kd in FLT3(ITD) but inhibition despite cell viability in FLT3-ITD was ~10-fold increased for 4b. Crenolanib and quizartinib lose affinity towards dual ITD and point-mutated FLT3, while affinity is increased for 4b. Especially towards the drug-resistant FLT3(ITD, F691L) mutation, 4b shows a ~15-fold increased affinity compared to crenolanib. Comparison of the affinity of 4b for the non-autoinhibited (Kd = 2.3 nM) and the autoinhibited state (Kd = 65 nM) of FLT3 shows a 28-fold decrease that is similar to type I inhibitors as sunitinib and midostaurin while type II inhibitors like sorafenib and quizartinib typically give a 100- to 1000-fold shift in affinity.37

Time-dependent inhibition of cell viability. To monitor the onset of cytotoxicity we compared crenolanib and 4b in a real-time cell viability assay with MV4-11 cells (Figure 4c, Figure S44). While cytotoxicity became detectable already after 2 h, robust fitting of cell viability IC50 was possible after 8 h for both compounds. Cytotoxicity by 4b relative to crenolanib was maximal at 18 h (4.4-fold) and plateaus to ~1.5x stronger inhibition for 4b after 24 h until the end of the assay (77 h). Interestingly, short incubation times (2 h, 8 h)


28

Page 29 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


followed by wash-out of the inhibitor and incubation in inhibitor-free medium negatively affects the performance of 4b compared to crenolanib (Figure 4d). However, potency of 4b is increased by incubation times longer than 24 h suggesting that rapid de novo synthesis of FLT3-ITD may lead to resistance to short incubation times of irreversible inhibitors. However, onset of inhibition by 4b demonstrated by decreased pFLT3 and downstream pSTAT5 levels is in the single-digit nanomolar range after 1 h (Figure 3a).

Evaluation of BODIPY-labeled Activity-based Probes. We further aimed to modify the hit compounds 2c and 2i with a boron-dipyrromethene (BODIPY) label attached to the solventexposed site of the fragment to image irreversible protein binding similar to a recently reported Ibrutinib-based probe38 (Figure 5a, b). Synthesis was performed similar to reported work.


29


Page 30 of 85

Figure 5. (a) Structure of fluorescent probes 5a and 5b and the reversible inhibitors 5c and 5d. (b) Binding model of 5a in FLT3. (c) Western blot of FLT3(D835Y) incubated with 5a or 5c (100 µM) and pre-treatment with IAA (10 mM) or 4b (100 µM) followed by incubation with 5a. (d) Binding of 5a – d (50 µM) to FLT3(D835Y) and pre-treatment with 4b and hypothemycin (HY) followed by incubation with 5a. (e) Binding of 5a (10 µM) to FLT3(D835Y) and pre-treatment with 3m, 3n, 4a and sunitinib (SU, each 100 µM) followed by incubation with 5a (10 µM).

Inhibition of FLT3(D835Y) by 5a (89%) and 5b (55%) was decreased compared to 4a and 4b (1 µM inhibitor) but still superior to the corresponding reversible inhibitors 5c (34%) and 5d (23%) (Figure S45). Incubation of FLT3(D835Y) with probes 5a - d and subsequent standard gel electrophoresis and Western blot allows fluorescent detection of covalent binding for 5a and 5b (50 µM, Figure 5c, d, Figure S46) while acetamide 5d shows no fluorescence. Interestingly, trifluoroacetamide 5c shows similar binding compared to 5b (18% at 100 µM and 7% at 50 µM; relative integrated density compared to 5a). We thus pretreated FLT3(D835Y) with reversibly and irreversibly acting inhibitors following incubation with fluorescent probe 5a that gave the highest fluorescence signal and signal to background ratio. Treatment of FLT3(D835Y) with 10 mM IAA almost completely abrogated fluorescence after treatment with 5a (100 µM). Incubation with 4b (100 µM) gave a small decrease of fluorescence with 100 µM of probe 5a (-28%) while incubation with 50 µM probe


30

Page 31 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


decreased fluorescence signal further (-41%) and comparable to treatment with hypothemycin (-52%). A similarly decreased fluorescence was observed after pre-incubation with bisaminopyrimidines 3m (-55%), 3n (-90%) and indolinone 4a (-68%). Incubation with the reversible inhibitor sunitinib (-29%) slightly decreases fluorescence suggesting covalent binding of the probe in the ATP site where the 10x excess of pre-treated inhibitor strongly competes with the reversible complex preceding covalent binding of the irreversible probe (Figure 5e). Additionally, gastrointestinal uptake of probes 5a and 5b could be detected in

Danio rerio embryos (Figure S47).

Conclusions and Perspective In summary, we have discovered irreversible binding FLT3 inhibitors that overcome drugresistant mutations of FLT3. A focused screening library was designed by attachment of chemically reactive groups to potent reversible binding fragments. Subsequent covalent docking identified two privileged scaffolds that gave numerous FLT3 inhibitors with improved cytotoxicity in FLT3-driven cell lines. Hybridization with the parental sunitinib scaffold gave the most potent compound 4b (BSc5371) that shows increased cytotoxicity in comparison to drugs in current clinical trials and is selective over closely related kinases. Subsequent development of an activity-based probe enabled imaging of covalent binding and competition with non-fluorescent inhibitors. The screening strategy presented here is adaptable to many drug targets and may serve as a blueprint for the rational in silico design of covalent drugs enabling more time- and cost-efficient drug discovery.


31


Page 32 of 85

Experimental Section Covalent Virtual Screening and Filter Strategy Homology Model The protein sequence of FLT3 (PDB: 1RJB) and the structure of cKIT in complex with ADP (PDB: 1PKG) was obtained from the protein databank. Amino acid residues that are only described in one of the models and are distant from the ATP-binding pocket were discarded (see Table S1 and Figure S1). The homology model was built using Molecular Operating Environment 2014.09 (MOE, Chemical Computing Group). Protein sequences were aligned and the sequence of 1RJB was modeled on the template 1PKG using 25 mainchain models and the Amber12:EHT forcefield. RMSD between the template and the final model is 0.01 Å. The Ramachandran Plot of the final model displays four geometric outliers that are all distant from the ATP-binding pocket and were not further optimized. Reversible Docking Initial pharmacophores for the 13 selected reversible inhibitor scaffolds were built using a superposition of reported crystal structures and the FLT3 DFG-in homology model to enable robust evaluation of the most likely reversible binding pose (Figures S6 to S18). Input structures were drawn using ChemBioDraw Ultra 14.0 and energy minimized using MOE and the MMFF94x forcefield. The final selected docking pose that serves as a template was used to define the pharmacophore for the covalent virtual screening.


32

Page 33 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Covalent Docking using Docktite All compounds studied were docked in the ATP-binding site where the reversible inhibitor serves as a site indicator and defines the pharmacophore. Covalent docking was performed using the Docktite SVL script set for MOE. Input structures were generated by modification of the reversible docked inhibitor scaffold with a chemical reactive group at an appropriate site to give 128 inhibitors (see figure S5). The chemical reactive groups are tagged and attached to the covalent attachment point to give 160 sidechain-attached ligands that served as ligands for covalent docking. Pharmacophores were generated automatically regarding the features of the covalent attachment point (R = 0.4 Å) and have been enhanced using aryl and hydrogen-bond donor and acceptor features (R = 1.5 Å, see figure S6B to S18B). The pharmacophore feature of the nucleophilic sulfur atom is increased to 3.0 Å to allow conformational flexibility of the attachment point, while the other heavy atoms of Cys828 are unchanged to resemble the initial input conformation of the homology model. The Cys828 residue and the N- and C-terminal amide group were removed from the receptor for covalent docking. Docking was performed using MOE´s rigid receptor docking function and the Amber12:EHT forcefield. 5000 conformations of the attached ligands are sampled and filtered by the pharmacophore model. Rescoring of the placements was performed using the London dG scoring function and 100 poses are subjected to grid-based minimization and rescored using London dG. The force constant of the pharmacophore restraint was set to 1e+9. The docking was performed in batch jobs and was finished after approximately 48 h using a custom-built 64-bit computer system (Intel i7-4790K CPU 4*4GHz, 8 GB Ram 2800


33


Page 34 of 85

MHz). The attached sidechain was then cleaved from the inhibitor using Docktite to allow more robust rescoring of the ligand. 831 poses have been retained and are subjected to the filter strategy. Filter strategy In the first step for scaffold selection, DSX rescoring was performed using the linux-based standalone version. Ligands that contain any of the three characteristic backbone interactions with the hinge region (gatekeeper +1 residue Glu692 backbone donor interaction; and/or gatekeeper +3 residue Cys694 backbone donor/acceptor interaction) are selected using the Protein-Ligand Interaction Fingerprints (PLIFs) function of MOE (404

poses remaining). From these poses only the top 5 poses of each ligand are retained (319 poses remaining). In the last filter step, ligands with a ligand efficiency of more than -0.5 HA/kcal are discarded (LEV = (London dG Score) / (number of heavy atoms). Redocking of the experimental validated hits was done like the screening method using forcefield minimization and an enhanced pharmacophore radius of the scaffold pharmacophore (R = 2Å). Molecular Dynamics The molecular dynamics simulations of 4b were performed using MOE2016.09 and the NPA algorithm39 starting from the top-scored docking pose of 4b in the FLT3 DFG-in homology model. 10 picoseconds of equilibration are followed by a 100 ps production step using a time step of 0.1 ps. All bond lengths have been constrained. Time step for the equation of motions


34

Page 35 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was set to 0.001. Temperature was set to 300K. The time-dependent RMSD values are given in the SI (Figure S31). Chemistry General remarks Solvents and chemicals were purchased from ABCR, Acros Organics, Alfa Aesar, Deutero Gmbh, Fluka, Grüssing Gmbh, Merck, Carl Roth, Sigma-Aldrich and VWR and were used without further purification unless stated otherwise. Qualitative analytical thin layer chromatography was performed using silica gel 60 plates with fluorescent indicator F254 from Merck. Developed plates were dried and observed under UV light (254 or 365 nm). Silica chromatography was performed using silica gel 60 (0.015 – 0.04 mm particle size) from Merck and an isocratic solvent mixture. Flash chromatography was performed using a Teledyne ISCO Combiflash Rf 4x system with pre-packed silica columns obtained from Telos. The column was conditioned with solvent to the starting concentration of the gradient. Peaks were detected at 214/254 nm (reversed-phase chromatography) or 254/280 nm (normal phase chromatography). 1H and 13C-NMR spectra were recorded on a Bruker AR 300 (1H at 300 MHz, 13C at 75 MHz) or a Bruker DRX 500 spectrometer (1H at 500 MHz, 13C at 126 MHz). The specific frequency and solvents are stated for each experiment. Chemical shifts δ are expressed in ppm using the solvent residual proton signal as reference. Signal splitting patterns are described as singlet (s), doublet (d), triplet (t), quartet (q), multiplet (m) or a combination thereof. Coupling constants J are rounded to one decimal place and given in


35


Page 36 of 85

Hz. The spectra were evaluated using MestReNova 11 by Mestrelab Research. High Performance Liquid Chromatography was performed with an Agilent 1100 system using a phenomenex synergi polar reversed phase column (4 µm particle size, 150 x 3.0 mm, pore size 80 Å) connected to a variable wavelength detector. Mobile phase consists of water/acetonitrile + 0.1% trifluoro acetic acid forming a linear gradient starting with 30% water (held for 1 minute) increased to 90% acetonitrile over 10 min and held for one min with a constant flow of 1 ml/min. Microwave assisted synthesis were performed in a Biotage Initiator plus microwave reactor using the conditions stated in the corresponding experiments. EI-MS spectra were recorded on a double focused mass spectrometer MAT95. ESI-MS was obtained using a Bruker Daltonics qTOF spectrometer. All compounds used in biochemical assays have a purity of more than 95% as determined by HPLC and are reported at the corresponding experiments.

N4-(5-methyl-1H-pyrazol-3-yl)-N2-phenylpyrimidine-2,4-diamine (1a). To a solution of 2-chloro-N-(5-methyl-1H-pyrazol-3-yl)pyrimidin-4-amine (150 mg, 0.72 mmol, 1.0 equiv) in n-butanol (4 ml) was added aniline (67 mg, 0.72 mmol, 1.0 equiv) and 37% hydrochloric acid (20 µl). The mixture was stirred for 24 h at 100°C, cooled to room temperature and kept at 4°C overnight. The precipitate was isolated by filtration, washed with

cyclohexane

and

dichloromethane

and

purified

by

reversed-phase

flash

chromatography (acetonitrile/H2O = 10 to 100%) to give 1a as a colorless amorphous solid. Yield: 115 mg, 0.432 mmol, 60%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 12.40 (s, 1H),


36

Page 37 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


11.03 (s, 1H), 10.65 (s, 1H), 8.00 (d, J = 7.1 Hz, 1H), 7.59 (d, J = 7.7 Hz, 2H), 7.44 (t, J = 7.9 Hz, 2H), 7.24 (t, J = 7.4 Hz, 1H), 6.54 (s, 1H), 6.22 (s, 1H), 2.20 (s, 3H).

13C

NMR (126 MHz,

DMSO-d6) δC (ppm) = 159.9, 159.4, 159.1, 153.1, 146.0, 139.1, 137.0, 128.9, 124.8, 117.9, 96.8, 10.6. MS (EI): m/z = 266 [M] +. HPLC (254 nm): tR = 1.827 min (99.4713%). (Z)-3-Benzylideneindolin-2-one (1b).40 To a solution of indolin-2-one (200 mg, 1.5 mmol, 1.0 equiv) and benzaldehyde (159 mg, 1.502 mmol, 1.0 equiv) in tetrahydrofuran (4 ml), piperidine (25 µl) was added and the mixture was heated in a microwave reactor at 120°C for 90 min. The solvent was evaporated, and the crude product purified by flash chromatography (dichloromethane/MeOH = 0 to 10%) to yield the product as a yellow amorphous solid. Yield: 122 mg, 0.55 mmol, 37%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 10.61 (s, 1H), 7.75 – 7.70 (m, 2H), 7.66 (s, 1H), 7.60 – 7.45 (m, 4H), 7.25 (td, J = 7.7, 1.2 Hz, 1H), 6.95 – 6.78 (m, 2H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 168.6, 143.0, 135.7,

134.5, 130.2, 129.6, 129.2, 128.7, 127.7, 122.3, 121.1, 120.9, 110.1. MS (ESI): m/z = 244.08 [M+Na] +. HPLC (254 nm): tR = 6.412 min (96.7002%). 3-Phenylimidazo[1,2-a]pyridine (1c).41 To a solution of imidazo[1,2-a]pyridine (100 mg, 0.846 mmol, 1.0 equiv) and bromobenzene (251 mg, 1.69 mmol, 2.0 equiv) in dry DMF (4 ml) was added Pd(OAc)2 (19 mg, 0.085 mmol, 0.1 equiv) and KOAc (166 mg, 1.69 mmol, 2.0 equiv) and the mixture was degassed under argon for 30 min. The reaction mixture was stirred in a microwave reactor at 160°C for 1 h. Then it was cooled to room temperature poured, in a mixture of saturated aqueous NaHCO3 solution (10 mL) and ethyl


37


Page 38 of 85

acetate (20 ml). The organic layer was washed with saturated aqueous sodium bicarbonate solution (2 x 10 ml) and water (3 x 10 ml), dried over MgSO4 and concentrated under reduced pressure. The crude product was purified by reversed-phase flash chromatography (acetonitrile/H2O = 10 to 100%) to give 1c as a colorless amorphous solid. Yield: 95 mg, 0.49 mmol, 58%. 1H NMR (500 MHz, CDCl3 + 10% DMSO-d6) δH (ppm) = 8.41 (d, J = 6.8 Hz, 1H), 8.15 (d, J = 9.0 Hz, 1H), 7.84 (s, 1H), 7.77 (t, J = 7.9 Hz, 1H), 7.59 – 7.46 (m, 5H), 7.29 (t, J = 6.9 Hz, 1H).

13C

NMR (126 MHz, CDCl3 + 10% DMSO-d6) δC (ppm) = 140.3, 132.5, 130.8,

129.9, 129.1, 127.3, 124.9, 124.6, 121.0, 117.3, 114.2. MS (EI): m/z = 194 [M] +. HPLC (254 nm): tR = 5.602 min (95.8210%). 6,7-Dimethoxy-N-phenylquinazolin-4-amine (1d). To a solution of 4-chloro-6,7dimethoxyquinazoline (50 mg, 0.22 mmol, 1.0 equiv) in isopropyl alcohol (2 ml) was added aniline (32 mg, 0.34 mmol, 1.55 equiv). The mixture was refluxed for 16 h and cooled to room temperature. The precipitate formed was isolated by filtration and washed with ethyl acetate (1 ml) twice to give the product as a colorless amorphous solid. Yield: 62 mg, 0.216 mmol, 98%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.51 (s, 1H), 8.81 (s, 1H), 8.40 (s, 1H), 7.79 – 7.66 (m, 2H), 7.50 (t, J = 7.9 Hz, 2H), 7.41 (s, 1H), 7.38 – 7.29 (m, 1H), 4.04 (s, 3H), 4.00 (s, 3H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 158.2, 156.2, 150.2, 148.6,

136.86, 135.4, 129.5, 128.7, 126.3, 124.9, 107.2, 104.1, 99.7, 57.0, 56.4. MS (EI): m/z = 281 [M] +.

HPLC (254 nm): tR = 5.997 min (97.2261%).


38

Page 39 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3,5-Diphenyl-1H-pyrrolo[2,3-b]pyridine (1e). 3,5-Diphenyl-1-tosyl-1H-pyrrolo[2,3b]-pyridine (56 mg, 0.13 mmol, 1.0 equiv) was dissolved in a mixture of MeOH (15 ml) and aqueous 4 N sodium hydroxide (0.2 ml) and stirred at 50°C for 24 h. The reaction mixture was cooled to room temperature and adjusted to pH 5-6 by addition of aqueous 2 N hydrochloric acid. The mixture was concentrated and treated with water (10 ml) and ethyl acetate (10 ml). The aqueous layer was extracted with ethyl acetate (10 ml) three times and the combined organic layers were washed with brine, dried over MgSO4 and the solvent was evaporated. The crude product was purified by silica chromatography (cyclohexane/ethyl acetate = 1/1) to give the title compound as a colorless amorphous solid. Yield: 26 mg, 0.095 mmol, 73%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.99 (s, 1H), 8.57 (d, J = 2.1 Hz, 1H), 8.43 (d, J = 2.0 Hz, 1H), 7.91 (d, J = 2.6 Hz, 1H), 7.78 (ddd, J = 12.7, 8.2, 1.1 Hz, 4H), 7.47 (ddd, J = 19.7, 10.9, 4.6 Hz, 4H), 7.41 – 7.35 (m, 1H), 7.30 – 7.23 (m, 1H).

13C

NMR (126

MHz, DMSO-d6) δC (ppm) = 148.6, 141.9, 139.0, 134.9, 128.9, 128.9, 128.8, 127.1, 126.9, 126.39, 125.7, 125.4, 124.6, 117.3, 114.7. MS (EI): m/z = 270 [M] +. HPLC (254 nm): tR = 2,653 min (97,4023%).

N4-(5-Cyclopropyl-1H-pyrazol-3-yl)-N2-phenylpyrimidine-2,4-diamine (1f). To a solution

of

2-chloro-N-(5-cyclopropyl-1H-pyrazol-3-yl)pyrimidin-4-amine

(200

mg,

0.849 mmol, 1.0 equiv) in n-butanol (3.2 ml) was added aniline (79 mg, 0.849 mmol, 1.0 equiv) and 37 % hydrochloric acid (40 µl). The mixture was stirred for 24 h at 100°C and subsequently cooled to room temperature and kept at 4°C overnight. The precipitate was


39


Page 40 of 85

isolated by filtration, washed with cyclohexane and dichloromethane and purified by reversed-phase flash chromatography (acetonitrile/H2O =

10 to 100%) to give 1f as a

colorless amorphous solid. Yield: 159 mg, 0.54 mmol, 64%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 12.48 (s, 1H), 11.20 (s, 1H), 10.68 (s, 1H), 8.21 – 7.81 (m, 1H), 7.56 – 7.47 (m, 2H), 7.44 (t, J = 7.7 Hz, 2H), 7.27 (d, J = 7.4 Hz, 1H), 6.46 (s, 1H), 6.04 (s, 1H), 1.83 (dt, J = 8.8, 3.8 Hz, 1H), 1.11 – 0.89 (m, 2H), 0.55 (s, 2H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 176.6, 159.7, 157.4, 152.7, 146.5, 145.7, 142.2, 136.5, 129.0, 125.4, 123.4, 99.2, 93.3, 7.9, 6.7. MS (ESI): m/z = 293.1 [M+H] +. HPLC (254 nm): tR = 1.778 min (98.6020%). (Z)-3-((1H-Pyrrol-2-yl)methylene)indolin-2-one (1g).40 To a solution of indolin-2one (200 mg, 1.5 mmol, 1.0 equiv) and 1H-pyrrole-2-carbaldehyde (142 mg, 1.502 mmol, 1.0 equiv) in tetrahydrofuran (4 ml), piperidine (25 µl) was added and the mixture was heated in a microwave reactor at 120°C for 90 min. The solvent was evaporated, and the crude product purified by flash chromatography (dichloromethane/MeOH = 0 to 10%) to yield the product as a yellow amorphous solid. Yield: 131 mg, 0.62 mmol, 42%. 1H NMR (300 MHz, DMSO-d6) δH (ppm) = 13.34 (s, 1H), 10.88 (s, 1H), 7.74 (s, 1H), 7.66 – 7.60 (m, 1H), 7.35 (td, J = 2.7, 1.4 Hz, 1H), 7.15 (td, J = 7.6, 1.2 Hz, 1H), 7.00 (td, J = 7.6, 1.1 Hz, 1H), 6.89 (dt, J = 7.6, 0.9 Hz, 1H), 6.84 (dt, J = 3.6, 1.7 Hz, 1H), 6.36 (dt, J = 3.7, 2.3 Hz, 1H). 13C NMR (75 MHz, DMSO-d6) δC (ppm) = 169.2, 138.9, 129.5, 126.8, 126.3, 125.6, 125.1, 121.2, 120.2, 118.5, 116.8, 111.4, 109.5. MS (EI): m/z = 210.0 [M] +. HPLC (254 nm): tR = 2.967 min (98.4849%).


40

Page 41 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)acetamide

(2a).

To

a

solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (100 mg, 0.444 mmol, 1.0 equiv) in THF (20 ml) at 0°C was added triethylamine (123 µl, 0.888 mmol, 2.0 equiv) and acetyl chloride (35 µl, 0.488 mmol, 1.1 equiv). The solution was warmed to room temperature and stirred for 2 h. The reaction was quenched by the addition of brine (20 ml) and ethyl acetate (20 ml) and the organic layer was washed with 0.1 N aqueous HCl (2 x 10 ml) and 0.1 N aqueous NaOH solution (2 x 10 ml). The organic layer was dried over Na2SO4, the solvent was removed, and the crude product was purified by flash chromatography (dichloromethane/MeOH = 1 to 10%) to give the product as a yellow amorphous solid. Yield: 69 mg, 0.257 mmol, 58%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.36 (s, 1H), 10.80 (s, 1H), 9.80 (s, 1H), 7.92 (d, J = 2.0 Hz, 1H), 7.60 (s, 1H), 7.37 (dt, J = 4.1, 2.0 Hz, 1H), 7.20 (d, J = 8.3, 2.0 Hz, 1H), 6.93 (dt, J = 3.5, 1.7 Hz, 1H), 6.83 (d, J = 8.3 Hz, 1H), 6.37 (dt, J = 3.6, 2.4 Hz, 1H), 2.05 (s, 3H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 169.3, 167.8, 134.9, 133.5, 129.5, 126.1, 125.7, 125.1, 120.6, 118.8, 117.0, 111.4, 110.5, 109.4, 40.1, 39.9, 39.7, 39.5, 39.4, 39.2, 39.1, 23.8. MS (ESI) m/z = 268.1 [M+H] +. HPLC (254 nm): tR = 3.843 min (100.0000%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)-2-fluoroacetamide (2b). 2d (124 mg, 0.355 mmol, 1.0 equiv) and potassium fluoride (206 mg, 3.55 mmol, 10.0 equiv) were suspended in diethylene glycol (5 ml) and stirred at 130°C for 5 h. The reaction mixture was cooled to room temperature and diluted with water (100 ml), extracted with dichloromethane (2 x 50 ml), dried over Na2SO4 and the solvent was evaporated. The residue


41


Page 42 of 85

was purified by silica chromatography (dichloromethane/MeOH=20/1) to give the title compound as a yellow amorphous solid. Yield: 35 mg, 0.123 mmol, 35%. 1H NMR (300 MHz, DMSO-d6) δH (ppm) = 13.34 (s, 1H), 10.86 (s, 1H), 9.97 (s, 1H), 7.93 (d, J = 2.0 Hz, 1H), 7.62 (s, 1H), 7.49 – 7.24 (m, 2H), 7.01 – 6.78 (m, 2H), 6.36 (dt, J = 3.8, 2.3 Hz, 1H), 5.06 (s, 1H), 4.90 (s, 1H).

13C

NMR (75 MHz, DMSO-d6) δC (ppm) = 169.3, 165.6, 165.4, 135.5, 131.9,

129.4, 126.3, 125.8, 125.1, 120.8, 119.7, 116.7, 111.4, 109.4, 81.2, 78.8. MS (ESI) m/z = 286.1 [M+H] +. HPLC (254 nm): tR = 4.282 min (95.1175%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)-2-chloroacetamide (2c). To a solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (106 mg, 0.47 mmol, 1.0 equiv) in THF (15 ml) at 0°C was added pyridine (57 µl, 0.71 mmol, 1.5 equiv) and chloro acetyl chloride (49 µl, 0.611 mmol, 1.3 equiv). The solution was warmed to room temperature and stirred for 1 h. The reaction was quenched by the addition of brine (20 ml) and ethyl acetate (20 ml) and the organic layer was washed with saturated aqueous sodium bicarbonate solution (20 ml) and brine (20 ml), dried with Na2SO4 and the solvent was evaporated.

The

residue

was

purified

by

silica

chromatography

(dichloromethane/MeOH=20/1) to give the title compound as a yellow amorphous solid. Yield: 53 mg, 0.18 mmol, 37%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.42 (s, 1H), 10.66 (s, 1H), 10.07 (s, 1H), 7.89 (d, J = 1.9 Hz, 1H), 7.47 (s, 1H), 7.23 – 7.17 (m, 2H), 6.83 – 6.76 (m, 2H), 6.31 (dt, J = 3.6, 2.3 Hz, 1H), 4.17 (s, 2H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) =


42

Page 43 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


169.3, 164.2, 135.5, 132.3, 129.5, 125.9, 125.2, 124.9, 120.1, 118.8, 117.0, 111.2, 110.5, 109.3, 43.3. MS (ESI) m/z = 302.1 [M+H] +. HPLC (254 nm): tR = 4.985 min (96.1842%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)-2-bromoacetamide (2d). To a solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (247 mg, 1.1 mmol, 1.0 equiv) in dry THF (20 ml) at 0°C was added diisopropylethylamine (382 µl, 2.2 mmol, 2.0 equiv) and bromo acetyl bromide (242 mg, 1.2 mmol, 1.1 equiv). The solution was warmed to room temperature and stirred for 1 h. The reaction was quenched by the addition of brine (20 ml) and ethyl acetate (40 ml), the organic layer was washed with saturated aqueous sodium bicarbonate solution (20 ml) and brine (20 ml), dried with Na2SO4 and the solvent was

evaporated.

The

residue

was

purified

by

silica

chromatography

(dichloromethane/MeOH=20/1) to give the title compound as a yellow amorphous solid. Yield: 163 mg, 0.472 mmol, 47%. 1H NMR (300 MHz, DMSO-d6) δH (ppm) = 13.34 (s, 1H), 10.86 (s, 1H), 10.27 (s, 1H), 7.93 (d, J = 2.0 Hz, 1H), 7.65 (s, 1H), 7.36 (q, J = 2.1 Hz, 1H), 7.20 (dd, J = 8.4, 2.0 Hz, 1H), 7.00 – 6.77 (m, 2H), 6.36 (dt, J = 4.1, 2.3 Hz, 1H), 4.04 (s, 2H). 13C NMR (75 MHz, DMSO-d6) δC (ppm) = 169.2, 164.4, 135.4, 132.6, 129.5, 126.4, 125.9, 125.2, 120.8, 118.9, 116.7, 111.5, 110.5, 109.6, 30.4. MS (ESI) m/z = 346.03 [M+H] +. HPLC (254 nm): tR = 5.145 min (95.9804%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)propionamide (2e). To a solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (57 mg, 0.253 mmol, 1.0 equiv) in acetonitrile (4 ml) was added triethylamine (35 µl, 0.235 mmol, 1.0 equiv), 1-(3-


43


Page 44 of 85

dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (97 mg, 0.51 mmol, 2.0 equiv) and propanoic acid (23 µl, 0.304 mmol, 1.2 eq). The reaction was stirred at room temperature overnight and water (20 ml) and ethyl acetate (20 ml) were added. The organic layer was washed with saturated aqueous sodium bicarbonate solution (20 ml) and brine (20 ml), dried over sodium sulfate and the solvent was evaporated. The residue was purified by silica chromatography (dichloromethane/MeOH=20/1) to give the title compound as a yellow amorphous solid. Yield: 21 mg, 0.075 mmol, 30%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.55 – 13.29 (m, 1H), 10.52 (s, 1H), 9.50 (s, 1H), 7.96 – 7.91 (m, 1H), 7.39 (s, 1H), 7.18 – 7.13 (m, 2H), 6.76 (d, J = 8.3 Hz, 1H), 6.72 (dt, J = 3.6, 1.7 Hz, 1H), 6.30 (dt, J = 3.7, 2.3 Hz, 1H), 2.33 (q, J = 7.6 Hz, 2H), 1.15 (t, J = 7.6 Hz, 3H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 171.8, 169.4, 134.9, 133.2, 129.5, 125.5, 125.0, 124.6, 119.7, 118.6, 117.3, 111.1, 110.4, 109.2, 29.5, 9.6. MS (EI) m/z = 281 [M] +. HPLC (254 nm): tR = 4.881 min (95.0225%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)acrylamide

(2f).

To

a

solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (153 mg, 0.679 mmol, 1.0 equiv) in dichloromethane (25 ml) at 0°C was added triethylamine (188 µl, 1.358 mmol, 2.0 equiv) and acryloyl chloride (60 µl, 0.75 mmol, 1.1 equiv). The solution was warmed to room temperature and stirred overnight. The reaction was quenched by the addition of water (20 ml) and ethyl acetate (20 ml) and the organic layer was washed with saturated aqueous sodium bicarbonate solution (2 x 20 ml) and brine (20 ml) and dried over Na2SO4. The solvent was evaporated, and the crude product was purified by silica chromatography


44

Page 45 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(dichloromethane/MeOH = 20/1) to give the title compound as a yellow amorphous solid. Yield: 65 mg, 0.232 mmol, 34%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.33 (s, 1H), 10.83 (s, 1H), 10.10 (s, 1H), 8.00 (d, J = 1.9 Hz, 1H), 7.63 (s, 1H), 7.45 – 7.24 (m, 2H), 7.02 – 6.80 (m, 2H), 6.45 (dd, J = 16.9, 10.1 Hz, 1H), 6.37 – 6.15 (m, 1H), 5.75 (dd, J = 10.1, 2.0 Hz, 1H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 169.3, 163.0, 135.2, 133.0, 131.9, 129.4,

126.4, 126.3, 125.9, 125.2, 120.8, 119.0, 116.7, 111.5, 110.6, 109.6. MS (ESI) m/z = 280.5 [M+H] +. HPLC (254 nm): tR = 4.752 min (95.2080%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)propionamide (2g). To a solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (115 mg, 0.51 mmol, 1.0 equiv) in acetonitrile (4 ml) at 0°C was added triethylamine (71 µl, 0.51 mmol, 1.0 equiv), 1-(3dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (195 mg, 1.02 mmol, 2.0 equiv) and propargylic acid (38 µl, 0.612 mmol, 1.2 equiv) and stirred overnight at room temperature. The reaction was quenched by the addition of brine (20 ml) and ethyl acetate (20 ml) and the organic phase was washed with 0.1 N aqueous HCl (2 x 10 ml) and 0.1 N aqueous NaOH (2 x 10 ml). The organic phase was dried over Na2SO4, the solvent was removed,

and

the

crude

product

was

purified

by

flash

chromatography

(dichloromethane/MeOH = 1 to 10%) to give the product as a yellow amorphous solid. Yield: 61 mg, 0.22 mmol, 43%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.42 (s, 1H), 10.72 (s, 1H), 10.54 (s, 1H), 7.89 (d, J = 2.0 Hz, 1H), 7.49 (s, 1H), 7.33 – 7.19 (m, 2H), 6.86 – 6.75 (m, 2H), 6.32 (dt, J = 3.7, 2.3 Hz, 1H), 4.02 (s, 1H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) =


45


Page 46 of 85

169.3, 149.4, 135.6, 132.2, 129.4, 126.0, 125.1, 125.1, 120.3, 119.0, 111.3, 110.7, 109.3, 75.9. MS (EI) m/z = 277 [M] +. HPLC (254 nm): tR = 4.639 min (95.7320%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)ethanesulfonamide (2h). To a solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (57 mg, 0.25 mmol, 1.0 equiv) in THF (5 ml) at 0°C was added diisopropyl ethyl amine (44 µl, 0.25 mmol, 1.0 equiv) and ethane sulfonyl chloride (24 µl, 0.25 mmol, 1.0 equiv). The solution was warmed to room temperature and stirred overnight. The reaction was quenched by the addition of brine (20 ml) and ethyl acetate (20 ml) and the organic layer was washed with saturated aqueous sodium bicarbonate solution (20 ml) and brine (20 ml) and dried with Na2SO4. The solvent was evaporated, and the crude product was purified by flash chromatography (dichloromethane/MeOH = 1 to 10%) to give the title compound as a yellow amorphous solid. Yield: 18 mg, 0.057 mmol, 23%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.45 (s, 1H), 10.63 (s, 1H), 9.33 (s, 1H), 7.61 (ddd, J = 11.9, 8.2, 1.4 Hz, 1H), 7.41 (d, J = 2.0 Hz, 1H), 7.19 (dt, J = 3.8, 1.8 Hz, 1H), 7.01 (dd, J = 8.2, 2.1 Hz, 1H), 6.83 – 6.76 (m, 2H), 6.32 (dt, J = 3.7, 2.4 Hz, 1H), 2.99 (q, J = 7.4 Hz, 2H), 1.29 (t, J = 7.4 Hz, 3H). 13C NMR (126 MHz, DMSO-

d6) δC (ppm) = 169.3, 136.1, 131.6, 131.4, 129.4, 128.3, 126.1, 124.9, 120.7, 120.2, 112.1, 111.3, 109.7, 44.5, 7.8. MS (EI) m/z = 317 [M] +. HPLC (254 nm): tR = 4.965 min (96.3844%). (Z)-N-(3-((1H-pyrrol-2-yl)methylene)-2-oxoindolin-5-yl)ethenesulfonamide (2i). To a solution of (Z)-3-((1H-pyrrol-2-yl)methylene)-5-aminoindolin-2-one (112 mg, 0.5 mmol, 1.0 equiv) in THF (5 ml) was added diisopropyl ethyl amine (175 µl, 1.0 mmol, 2.0 equiv) and 2-


46

Page 47 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


chloroethane sulfonylchloride (53 µl, 0.5 mmol, 1.1 equiv). The reaction was stirred overnight at room temperature and quenched by the addition of brine (20 ml) and ethyl acetate (20 ml). The organic layer was dried over Na2SO4, the solvent was evaporated, and the crude product was purified by silica chromatography (dichloromethane/MeOH = 20/1) to give the title compound as a yellow amorphous solid. Yield: 57 mg, 0.18 mmol, 36%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.42 (s, 1H), 10.69 (s, 1H), 7.51 (s, 1H), 7.37 (d, J = 2.0 Hz, 1H), 7.22 (td, J = 2.6, 1.4 Hz, 1H), 6.96 (dd, J = 8.2, 2.1 Hz, 1H), 6.83 (d, J = 1.9 Hz, 1H), 6.79 (d, J = 8.2 Hz, 1H), 6.32 (d, J = 3.7 Hz, 1H), 6.04 (d, J = 16.6 Hz, 1H), 5.90 (d, J = 10.0 Hz, 1H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 169.3, 136.4, 136.0, 131.0, 129.5, 126.5, 126.3, 125.6, 125.1, 121.5, 120.4, 116.6, 113.0, 111.3, 109.6. MS (EI) m/z = 315 [M] +. HPLC (254 nm): tR = 5.227 min (96.8706%).

N-(3-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)acetamide (3a). To a solution of N2-(3-aminophenyl)-N4-(5-cyclopropyl-1Hpyrazol-3-yl)pyrimidine-2,4-diamine (129 mg, 0.42 mmol, 1.0 equiv) in ethyl acetate (15 ml) at

0°C was added acetic anhydride (79.4 µl, 0.84 mmol, 2.0 equiv) and the mixture was stirred at room temperature overnight. The precipitate was isolated by filtration and dried under high vacuum to give the title compound as a yellow amorphous solid. Yield: 78 mg, 0.22 mmol, 52%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.99 (s, 1H), 9.83 (s, 1H), 9.44 (d, J = 85.0 Hz, 2H), 9.35 – 8.90 (m, 1H), 8.00 (d, J = 5.8 Hz, 1H), 7.95 – 7.72 (m, 1H), 7.59 – 7.38 (m, 1H), 7.30 – 6.94 (m, 2H), 6.38 (s, 1H), 2.06 (s, 3H), 1.86 (tt, J = 8.4, 5.0 Hz, 1H), 0.91 (dq, J = 6.4,


47


Page 48 of 85

4.0 Hz, 2H), 0.80 – 0.63 (m, 2H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 168.09, 159.54, 155.74, 141.09, 139.35, 128.34, 114.29, 112.32, 110.13, 98.25, 59.72, 23.98, 20.72, 14.06, 8.52, 7.63. MS (ESI): m/z = 350.18 [M+H] +. HPLC (254 nm): tR = 1.514 min (95.6875%).

N-(4-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)acetamid (3b). To a solution of N2-(4-aminophenyl)-N4-(5-cyclopropyl-1Hpyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in dry THF (15 ml) at 0°C

was added pyridine (58 µl, 0.716 mmol, 1.1 equiv) and acetic acid anhydride (68 µl, 0.716 mmol, 1.0 equiv) and the solution was stirred at room temperature overnight. The solvent was evaporated, and the crude product was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a yellow amorphous solid. Yield: 124 mg, 0.355 mmol, 55%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 12.05 (s, 1H), 9.78 (s, 1H), 9.57 (s, 1H), 9.05 (s, 1H), 7.97 (d, J = 5.7 Hz, 1H), 7.72 – 7.59 (m, 2H), 7.55 – 7.42 (m, 2H), 6.38 (s, 1H), 6.08 (s, 1H), 2.04 (s, 3H), 1.87 (tt, J = 8.4, 5.1 Hz, 1H), 0.98 – 0.82 (m, 2H), 0.76 – 0.59 (m, 2H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 167.7, 159.6, 159.5,

155.9, 155.8, 136.2, 133.1, 119.3, 97.8, 59.7, 54.9, 23.8, 21.0, 20.7, 14.1, 7.7. MS (ESI): m/z = 350.18 [M+H] +. HPLC (254 nm): tR = 1.352 min (98.1651%). 2-Chloro-N-(3-((4-((5-cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)acet-amide (3c). To a stirred solution of N2-(3-aminophenyl)-N4-(5-cyclopropyl1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF (5 ml) was

added triethylamine (99 µl, 0.716 mmol, 1.1 equiv) and 2-chloroacetyl chloride (57 µl, 0.716


48

Page 49 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


mmol, 1.1 equiv) in THF (2 ml). After one hour additional triethylamine (36 µl, 0.26 mmol, 0.4 equiv) and 2-chloroacetyl chloride (21µl, 0.26 mmol, 0.4 equiv) was added and the solution was stirred overnight. Saturated aqueous sodium hydrogen carbonate solution (15 ml) and brine (15 ml) were added and the mixture was extracted with ethyl acetate (40 ml). The organic layer was dried over Na2SO4, the solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a yellow amorphous solid. Yield: 100 mg, 0.261 mmol, 40%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.97 (s, 1H), 10.21 (s, 1H), 9.48 (s, 1H), 9.09 (s, 1H), 7.93 (d, J = 71.4 Hz, 2H), 7.55 (d, J = 8.0 Hz, 1H), 7.23 (q, J = 11.5, 8.0 Hz, 2H), 6.38 (d, J = 67.1 Hz, 2H), 4.27 (s, 2H), 1.87 (tt, J = 8.5, 5.1 Hz, 1H), 0.92 (s, 2H), 0.78 – 0.58 (m, 2H).

13C

NMR (126

MHz, DMSO-d6) δC (ppm) = 164.4, 159.5, 155.6, 148.2, 145.4, 141.3, 138.5, 128.5, 115.2, 112.5, 110.6, 98.4, 93.0, 43.6, 7.6, 6.9. MS (ESI): m/z = 384.14 [M+H] +. HPLC (254 nm): tR = 2.180 min (97.3103%). 2-Chloro-N-(4-((4-((5-cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)acet-amide (3d). To a stirred solution of N2-(4-aminophenyl)-N4-(5-cyclopropyl1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF (15 ml) was

added triethylamine (99 µl, 0.716 mmol, 1.1 equiv) and 2-chloroacetyl chloride (57 µl, 0.716 mmol, 1.1 equiv) and the solution was stirred overnight. Saturated aqueous sodium hydrogen carbonate solution (15 ml) and brine (15 ml) were added and the mixture was extracted with ethyl acetate (40 ml). The organic phase was dried with sodium sulfate, the


49


Page 50 of 85

solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a yellow amorphous solid. Yield: 95 mg, 0.248 mmol, 38%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 12.00 (s, 1H), 10.18 (s, 1H), 9.49 (s, 1H), 9.04 (s, 1H), 8.11 – 7.85 (m, 1H), 7.70 (d, J = 8.6 Hz, 2H), 7.56 – 7.47 (m, 2H), 6.47 (d, J = 25.5 Hz, 1H), 6.33 – 6.11 (m, 1H), 4.24 (s, 2H), 1.87 (tt, J = 8.5, 5.1 Hz, 1H), 0.93 (d, J = 8.2 Hz, 2H), 0.69 (tt, J = 6.9, 4.4 Hz, 2H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 164.0, 159.4, 155.6, 136.96, 132.0, 119.7, 97.9, 43.6, 7.7. MS (ESI): m/z = 384.14 [M+H] +. HPLC (254 nm): tR = 1.949 min (97.8748%).

N-(3-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)propionamide (3e). To a solution of N2-(3-aminophenyl)-N4-(5-cyclopropyl-1Hpyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in dry THF (15 ml) was

added pyridine (63 µl, 0.781 mmol, 1.2 equiv) and propionic acid anhydride (100 µl, 0.781 mmol, 1.2 equiv) and the solution was stirred at room temperature overnight. The solvent was evaporated, the residue was dissolved in ethyl acetate and washed with water (30 ml). The organic extract was dried over Na2SO4 and the solvent was removed. The crude product was purified by flash chromatography (dichloromethane/MeOH = 0 to 10%) to yield the product as a yellow amorphous solid. Yield: 111 mg, 0.305 mmol, 47%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.97 (s, 1H), 9.76 (s, 1H), 9.49 (s, 1H), 9.04 (s, 1H), 8.00 (d, J = 5.6 Hz, 1H), 7.86 (s, 1H), 7.59 – 7.38 (m, 1H), 7.26 (d, J = 8.2 Hz, 1H), 7.17 (t, J = 8.0 Hz, 1H), 6.37 (s, 2H), 2.34 (q, J = 7.6 Hz, 2H), 1.86 (tt, J = 8.4, 5.0 Hz, 1H), 1.10 (t, J = 7.5 Hz, 3H), 0.91


50

Page 51 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(d, J = 8.1 Hz, 2H), 0.74 – 0.64 (m, 2H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 171.8,

155.6, 141.1, 139.4, 128.3, 112.4, 98.2, 54.9, 40.0, 39.9, 39.7, 39.5, 39.4, 39.2, 39.0, 29.5, 9.7, 7.6. MS (ESI): m/z = 364.19 [M+H] +. HPLC (254 nm): tR = 1.922 min (96.0849%).

N-(4-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)propionamide (3f). To a solution of N2-(4-aminophenyl)-N4-(5-cyclopropyl-1Hpyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1 equiv) in dry THF (15 ml) was

added pyridine (58 µl, 0.716 mmol, 1.1 equiv) and propionic acid anhydride (91 µl, 0.716 mmol, 1.1 equiv) and the solution was stirred at room temperature overnight. The solvent was evaporated, and the residue was dissolved in ethyl acetate and washed with water (30 ml). The organic extract was dried over Na2SO4 and the solvent was removed. The crude product was purified by flash chromatography (dichloromethane/MeOH = 0 to 10%) to yield the product as a yellow amorphous solid. Yield: 104 mg, 0.286 mmol, 40%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 12.17 (s, 1H), 10.00 (d, J = 15.7 Hz, 1H), 9.82 (s, 1H), 9.41 (s, 1H), 7.95 (d, J = 6.2 Hz, 1H), 7.57 (s, 4H), 6.42 (s, 1H), 6.05 (s, 1H), 2.33 (q, J = 7.6 Hz, 2H), 1.86 (tt, J = 8.4, 5.0 Hz, 1H), 1.11 (t, J = 7.6 Hz, 3H), 0.99 – 0.86 (m, 2H), 0.65 (dt, J = 6.6, 3.2 Hz, 2H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 171.5, 159.5, 157.7, 134.4, 120.9, 119.3, 98.1, 51.4, 29.5, 28.7, 28.5, 28.4, 9.8, 7.7, 7.1. MS (ESI): m/z = 364.20 [M+H] +. HPLC (254 nm): tR = 1.674 min (95.2979%).

N-(3-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)acrylamide (3g). To a stirred solution of N2-(3-aminophenyl)-N4-(5-cyclopropyl-


51


Page 52 of 85

1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF (15 ml) was

added pyridine (58 µl, 0.716 mmol, 1.1 equiv) and acryloyl chloride (59 µl, 0.716 mmol, 1.1 equiv) and the solution was stirred for one hour at 0°C and at room temperature overnight. The solvent was evaporated, and the residue was taken up in ethyl acetate and washed with saturated aqueous sodium hydrogen carbonate solution (15 ml) and brine (15 ml). The organic layer was dried over Na2SO4, the solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a colorless amorphous solid. Yield: 62 mg, 0.171 mmol, 26%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.97 (s, 1H), 10.06 (s, 1H), 9.46 (d, J = 68.4 Hz, 1H), 9.35 – 9.00 (m, 1H), 8.10 – 7.88 (m, 3H), 7.50 (dd, J = 8.2, 2.0 Hz, 1H), 7.37 (d, J = 8.2 Hz, 1H), 7.21 (t, J = 8.1 Hz, 1H), 6.50 (dd, J = 17.0, 10.2 Hz, 1H), 6.27 (dd, J = 16.9, 2.0 Hz, 1H), 5.75 (dd, J = 10.2, 2.1 Hz, 1H), 1.86 (tt, J = 8.4, 5.1 Hz, 1H), 0.90 (tt, J = 9.3, 5.4 Hz, 2H), 0.78 – 0.59 (m, 2H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 163.0, 159.4, 141.1, 139.1, 132.1, 128.5, 126.5, 112.7, 98.3, 7.6. MS (ESI): m/z = 362.18 [M+H] +. HPLC (254 nm): tR = 1.933 min (95.0714%).

N-(4-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)acrylamide (3h). To a stirred solution of N2-(4-aminophenyl)-N4-(5-cyclopropyl1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF (15 ml) was

added pyridine (58 µl, 0.716 mmol, 1.1 equiv) and acryloyl chloride (59 µl, 0.716 mmol, 1.1 equiv) and the solution was stirred for one hour at 0°C. The solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the


52

Page 53 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


title compound as a colorless amorphous solid. Yield: 30 mg, 0.083 mmol, 13%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.99 (s, 1H), 10.02 (s, 1H), 9.47 (s, 1H), 9.00 (s, 1H), 7.98 (d, J = 5.8 Hz, 1H), 7.70 (d, J = 8.6 Hz, 2H), 7.64 – 7.53 (m, 2H), 6.45 (dd, J = 16.9, 10.2 Hz, 1H), 6.25 (dd, J = 17.0, 2.1 Hz, 1H), 5.73 (dd, J = 10.2, 2.0 Hz, 1H), 1.87 (tt, J = 8.4, 5.0 Hz, 1H), 0.99 – 0.86 (m, 2H), 0.76 – 0.63 (m, 2H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 162.6, 159.5,

155.9, 136.7, 132.6, 132.1, 126.1, 119.6, 119.3, 97.9, 14.1, 7.7. MS (ESI): m/z = 362.18 [M+H] +. HPLC (254 nm): tR = 1.703 min (96.9724%).

N-(3-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)propiolamide (3i). To a stirred solution of N2-(3-aminophenyl)-N4-(5cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in DMF

(5 ml) at 0°C was added diisopropylethylamine (233 µl, 1.37 mmol, 2.1 equiv), propargylic acid (241 µl, 3.9 mmol, 6.0 equiv) and N-(3-dimethylaminopropyl)-N′-ethyl carbodiimide hydrochloride (263 mg, 1.37 mmol, 2.1 equiv) and the solution was stirred for two hours at 0°C. The reaction mixture was diluted with ethyl acetate and the solvent was removed by freeze drying. The residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a colorless amorphous solid. Yield: 12 mg, 0.033 mmol, 5%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.99 (s, 1H), 9.47 (s, 1H), 9.05 (s, 1H), 7.97 (s, 2H), 7.70 (d, J = 8.6 Hz, 2H), 7.55 – 7.45 (m, 2H), 6.56 – 6.08 (m, 2H), 4.34 (s, 1H), 1.87 (tt,

J = 8.5, 5.1 Hz, 1H), 0.93 (d, J = 8.1 Hz, 2H), 0.74 – 0.65 (m, 2H). 13C NMR (126 MHz, DMSO-


53


Page 54 of 85

d6) δC (ppm) = 159.9, 157.1, 153.5, 146.7, 135.0, 129.3, 117.6, 116.7, 116.7, 95.5, 76.2, 74.2, 5.3. MS (ESI): m/z = 360.15 [M+H] +. HPLC (254 nm): tR = 1.854 min (97.5080%).

N-(4-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)propiolamide (3j). To a stirred solution of N2-(4-aminophenyl)-N4-(5cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (130 mg, 0.42 mmol, 1.0 equiv) in DMF

(10 ml) at 0°C was added diisopropylethylamine (150 µl, 0.882 mmol, 2.1 equiv), propargylic acid (150 µl, 2.52 mmol, 6.0 equiv) and N-(3-dimethylaminopropyl)-N′-ethyl carbodiimide hydrochloride (169 mg, 0.882 mmol, 2.1 equiv) and the solution was stirred for two hours at 0°C. The reaction mixture was diluted with ethyl acetate and washed with 0.1 N HCl (20 ml) and 0.1 N NaOH (20 ml) and the organic layer was dried with sodium sulfate and the solvent removed by freeze-drying. The residue was purified by silica chromatography (DCM/MeOH = 10/1) to give the title compound as a colorless amorphous solid. Yield: 26 mg, 0.072 mmol, 17%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.97 (s, 1H), 10.71 (s, 1H), 9.49 (s, 1H), 9.06 (d, J = 39.8 Hz, 1H), 7.95 (d, J = 39.2 Hz, 2H), 7.56 (s, 1H), 7.21 (d, J = 5.1 Hz, 2H), 6.56 – 6.16 (m, 2H), 4.37 (s, 1H), 1.86 (tt, J = 8.4, 5.0 Hz, 1H), 0.96 – 0.85 (m, 2H), 0.74 – 0.64 (m, 2H).

13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 159.5, 155.5, 149.5, 141.2, 138.2, 128.5,

113.0, 98.3, 78.5, 76.8, 56.0, 18.5, 7.6. MS (ESI): m/z = 360.16 [M+H] +. HPLC (254 nm): tR = 1.692 min (99.0749%).

N-(3-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)ethanesulfon-amide (3k). To a solution of N2-(3-aminophenyl)-N4-(5-


54

Page 55 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF

(10 ml) was added triethylamine (180 µl, 1.3 mmol, 2.0 equiv) and ethane sulfonyl chloride (123 µl, 1.3 mmol, 2.0 equiv) and the mixture was stirred at room temperature for 1 h and at 60°C for 1 h. The reaction mixture was cooled to room temperature, diluted with ethyl acetate and washed with brine and saturated aqueous sodium bicarbonate solution. The organic layer was dried over Na2SO4, the solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a yellow amorphous solid. Yield: 81 mg, 0.203 mmol, 31%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.98 (s, 1H), 9.63 (s, 1H), 9.41 (s, 1H), 9.08 (s, 1H), 7.98 (s, 1H), 7.69 – 7.44 (m, 2H), 7.20 (t, J = 8.1 Hz, 1H), 6.86 – 6.74 (m, 1H), 6.46 (s, 1H), 6.30 (s, 1H), 3.12 (q, J = 7.3 Hz, 2H), 1.94 – 1.84 (m, 1H), 1.23 (t, J = 7.4 Hz, 3H), 0.98 – 0.85 (m, 2H), 0.71 (s, 2H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 159.7, 159.5, 155.8, 148.1, 145.4, 141.8, 138.5, 128.9, 114.8, 112.4, 110.6, 98.4, 93.05, 54.9, 45.0, 7.99, 7.7, 6.8. MS (ESI): m/z = 400.16 [M+H] +. HPLC (254 nm): tR = 2.142 min (96.4205%).

N-(4-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)ethanesulfon-amide (3l). To a solution of N2-(4-aminophenyl)-N4-(5cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF

(10 ml) was added triethylamine (180 µl, 1.3 mmol, 2.0 equiv) and ethane sulfonyl chloride (123 µl, 1.3 mmol, 2.0 equiv) and the mixture was stirred at room temperature for 1 h and at 60°C for 1 h. The reaction mixture was cooled to room temperature, diluted with ethyl


55


Page 56 of 85

acetate and washed with brine and saturated aqueous sodium bicarbonate solution. The organic layer was dried over Na2SO4, the solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a yellow amorphous solid. Yield: 102 mg, 0.256 mmol, 39%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.99 (s, 1H), 9.46 (s, 2H), 9.02 (s, 1H), 8.13 – 7.91 (m, 1H), 7.71 (d, J = 8.3 Hz, 2H), 7.15 (d, J = 8.4 Hz, 2H), 6.49 (d, J = 39.7 Hz, 1H), 6.22 (s, 1H), 3.03 (q, J = 7.3 Hz, 2H), 1.88 (dq, J = 8.8, 5.0, 4.3 Hz, 1H), 1.23 (t, J = 7.3 Hz, 3H), 1.00 – 0.86 (m, 2H), 0.77 – 0.64 (m, 2H). 13C

NMR (126 MHz, DMSO-d6) δC (ppm) = 159.5, 155.9, 148.1, 145.5, 137.7, 131.2, 121.3,

119.8, 98.0, 93.0, 54.9, 44.6, 8.0, 7.7, 6.8. MS (ESI): m/z = 400.16 [M+H] +. HPLC (254 nm): tR = 1.966 min (95.4217%).

N-(3-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)ethenesulfon-amide (3m). To a stirred solution of N2-(3-aminophenyl)-N4-(5cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (200 mg, 0.651 mmol, 1.0 equiv) in THF

(10 ml) was added triethylamine (90 µl, 0.65 mmol, 1.0 equiv) and 2-chloroethanesulfonyl chloride (68 µl, 0.651 mmol, 1.0 equiv) and the solution was stirred for one hour at 0°C. Additional triethylamine (90 µl, 0.65 mmol, 1.0 equiv) was added and the mixture was stirred at room temperature overnight. The reaction mixture was diluted with ethyl acetate and washed with saturated aqueous sodium bicarbonate solution (15 ml) and brine (15 ml). The organic layer was dried over Na2SO4, the solvent was removed, and the residue was purified by flash chromatography (dichloromethane/MeOH = 100/1 to 10/1) to give the title


56

Page 57 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


compound as a yellow amorphous solid. Yield: 40 mg, 0.1 mmol, 15%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 11.98 (s, 1H), 9.88 (d, J = 45.6 Hz, 1H), 9.50 (d, J = 85.1 Hz, 1H), 9.08 (s, 1H), 8.13 – 7.93 (m, 1H), 7.58 (d, J = 8.7 Hz, 2H), 7.16 (t, J = 7.7 Hz, 1H), 6.84 – 6.77 (m, 1H), 6.76 – 6.71 (m, 1H), 6.47 (s, 1H), 6.30 (s, 1H), 6.18 (d, J = 16.4 Hz, 1H), 6.06 (d, J = 9.9 Hz, 1H), 1.88 (s, 1H), 0.94 (d, J = 8.2 Hz, 2H), 0.70 (t, J = 13.8 Hz, 2H).

13C

NMR (126 MHz,

DMSO-d6) δC (ppm) = 161.0, 156.9, 149.4, 146.7, 143.0, 139.1, 137.6, 130.0, 128.8, 116.1, 113.7, 112.0, 99.8, 94.4, 56.2, 9.0, 8.1. MS (ESI): m/z = 398.15 [M+H] +. HPLC (254 nm): tR = 2.261 min (97.3760%).

N-(4-((4-((5-Cyclopropyl-1H-pyrazol-3-yl)amino)pyrimidin-2yl)amino)phenyl)ethenesulfon-amide (3n). To a stirred solution of N2-(4-aminophenyl)-N4-(5cyclopropyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (220 mg, 0.72 mmol, 1.0 equiv) in THF

(10 ml) at 0°C was added triethylamine (119 µl, 0.86 mmol, 1.2 equiv) and 2chloroethanesulfonyl chloride (90 µl, 0.86 mmol, 1.2 equiv) and the solution was stirred at room temperature overnight. Additional triethylamine (198 µl, 1.43 mmol, 2.0 equiv) and 2chloroethanesulfonyl chloride (75 µl, 0.72 mmol, 1.0 equiv) were added and the mixture was stirred at room temperature overnight. The reaction mixture was diluted with ethyl acetate and washed with saturated aqueous sodium hydrogen carbonate solution (15 ml) and brine (15 ml). The organic phase was dried over Na2SO4, the solvent was removed, and the residue was purified by silica chromatography (dichloromethane/MeOH = 10/1) to give the title compound as a yellow amorphous solid. Yield: 24 mg, 0.06 mmol, 9%. 1H NMR (500 MHz,


57


Page 58 of 85

CD3OD) δH (ppm) = 7.82 (d, J = 6.0 Hz, 1H), 7.52 – 7.39 (m, 2H), 7.09 – 7.00 (m, 2H), 6.56 (dd,

J = 16.5, 10.0 Hz, 1H), 6.17 (s, 1H), 6.02 (s, 1H), 5.83 (d, J = 9.9 Hz, 1H), 1.79 (tt, J = 8.5, 5.1 Hz, 1H), 0.91 – 0.76 (m, 2H), 0.61 (dq, J = 9.0, 3.4, 2.6 Hz, 2H). 13C NMR (126 MHz, CD3OD) δC (ppm) = 161.2, 156.9, 139.0, 137.2, 132.9, 127.7, 123.7, 123.0, 122.0, 117.4, 99.0, 8.3. MS (ESI): m/z = 398.15 [M+H] +. HPLC (254 nm): tR = 2.131 min (97.6470%). (Z)-5-((5-(2-Chloroacetamido)-2-oxoindolin-3-ylidene)methyl)-N-(2(diethylamino)ethyl)-2,4-dimethyl-1H-pyrrole-3-carboxamide (4a). To a solution of (Z)-5((5-amino-2-oxoindolin-3-ylidene)methyl)-N-(2-(diethylamino)ethyl)-2,4-dimethyl-1Hpyrrole-3-carboxamide (37 mg, 0.093 mmol, 1.0 equiv) in dry THF (5 ml) at 0°C was added diisopropyl ethylamine (32 µl, 0.186 mmol, 2.0 equiv) and 2-chloro acetyl chloride (8.9 µl, 0.11 mmol, 1.2 equiv). The reaction mixture was stirred for 18 h and quenched with brine (30 ml) and ethyl acetate (30 ml) and extracted with ethyl acetate (2 x 30 ml). The organic layers were combined, dried over Na2SO4 and the solvent was evaporated. The crude product was purified by reversed-phase flash chromatography (acetonitrile/H2O = 10 to 100%) to give the title compound as a yellow amorphous solid. Yield: 17 mg, 0.036 mmol, 39%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.60 (s, 1H), 10.83 (s, 1H), 10.12 (s, 1H), 7.84 (s, 1H), 7.46 (s, 1H), 7.38 (t, J = 5.6 Hz, 1H), 7.23 (d, J = 8.5 Hz, 1H), 6.82 (d, J = 8.3 Hz, 1H), 4.21 (s, 2H), 2.55 – 2.43 (m, 8H), 2.41 (s, 3H), 2.37 (s, 3H), 0.95 (t, J = 7.1 Hz, 6H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 169.5, 164.5, 164.3, 135.9, 135.1, 132.3, 129.3, 125.6, 125.6, 123.3,


58

Page 59 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


119.0, 111.1, 109.4, 51.7, 46.5, 43.5, 37.0, 13.3, 11.8, 10.5. MS (ESI) m/z = 472.2 [M+H] +. HPLC (254 nm): tR = 2.872 min (100.0000%).

(Z)-N-(2-(Diethylamino)ethyl)-2,4-dimethyl-5-((2-oxo-5-(vinylsulfonamido)indolin3-ylidene)-methyl)-1H-pyrrole-3-carboxamide (4b). To a solution of (Z)-5-((5-amino-2oxoindolin-3-ylidene)methyl)-N-(2-(diethylamino)ethyl)-2,4-dimethyl-1H-pyrrole-3carboxamide (68 mg, 0.17 mmol, 1.0 equiv) in dry THF (10 ml) at 0°C was added triethylamine (46 µl, 0.34 mmol, 2.0 equiv) and 2-chloroethanesulfonyl chloride (36 µl, 0.34 mmol, 2.0 equiv). The reaction mixture was stirred for 18 h and quenched with brine (20 ml) and ethyl acetate (20 ml) and extracted with ethyl acetate (2 x 20 ml). The organic layers were combined, dried over Na2SO4 and the solvent was evaporated. The crude product was purified by reversed-phase flash chromatography (acetonitrile/H2O = 10 to 100%) to give the title compound as a yellow amorphous solid. Yield: 34 mg, 0.07 mmol, 41%. 1H-NMR (500 MHz, CDCl3 / DMSO-d6 = 8/1) δH (ppm) = 13.35 (s, 1H), 10.02 (s, 1H), 7.08 (d, J = 2.0 Hz, 1H), 7.04 (s, 1H), 6.71 (dd, J = 8.3, 2.1 Hz, 1H), 6.51 (dd, J = 11.5, 7.0 Hz, 2H), 6.31 (dd, J = 16.6, 10.0 Hz, 1H), 5.83 (d, J = 16.7 Hz, 1H), 5.59 (d, J = 10.0 Hz, 1H), 3.23 – 3.18 (m, 2H), 2.45 (t, J = 6.1 Hz, 2H), 2.41 – 2.35 (m, 4H), 2.25 (s, 3H), 2.17 (s, 3H), 0.79 (d, J = 7.1 Hz, 6H). 13C

NMR (126 MHz, CDCl3 / DMSO-d6 = 8/1) δC (ppm) = 170.0, 165.5, 137.1, 136.20 135.6,

130.8, 128.9, 126.4, 126.3, 126.0, 123.0, 121.6, 119.2, 115.2, 113.2, 109.7, 51.4, 46.4, 36.5, 13.8, 11.3, 11.1. MS (ESI) m/z = 486.2 [M+H] +. HPLC (254 nm): tR = 2.862 min (96.1594%).


59


Page 60 of 85

(Z)-5-((5-Acetamido-2-oxoindolin-3-ylidene)methyl)-N-(2-(diethylamino)ethyl)-2,4dimethyl-1H-pyrrole-3-carboxamide (4c). To a solution of (Z)-5-((5-amino-2-oxoindolin-3ylidene)methyl)-N-(2-(diethylamino)ethyl)-2,4-dimethyl-1H-pyrrole-3-carboxamide (71 mg, 0.18 mmol, 1.0 equiv) in dry THF (10 ml) at 0°C was added triethylamine (50 µl, 0.36 mmol, 2 eq.) and acetyl chloride (26 µl, 0.36 mmol, 2.0 equiv). The reaction mixture was stirred for 18 h at room temperature and then quenched with brine (30 ml) and ethyl acetate (30 ml) and extracted with ethyl acetate (2 x 30 ml). The organic layers were combined, dried over Na2SO4 and the solvent was evaporated. The crude product was purified by reversed-phase flash chromatography (acetonitrile/H2O = 10 to 100%) to give the title compound as a yellow amorphous solid. Yield: 48 mg, 0.11 mmol, 61%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.64 (s, 1H), 10.81 (s, 1H), 9.77 (s, 1H), 8.05 (t, J = 5.4 Hz, 1H), 7.83 (d, J = 2.1 Hz, 1H), 7.67 (s, 1H), 7.54 (t, J = 5.3 Hz, 1H), 7.44 (s, 1H), 7.27 (dd, J = 8.3, 2.0 Hz, 1H), 7.07 (d, J = 4.0 Hz, 1H), 6.83 (d, J = 8.3 Hz, 1H), 6.37 (d, J = 4.1 Hz, 1H), 6.28 (s, 1H), 3.30 – 3.25 (m, 1H), 3.12 (t, J = 7.7 Hz, 2H), 2.48 (s, 3H), 2.46 (s, 3H), 2.40 (s, 3H), 2.26 (s, 3H), 2.04 (s, 3H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 169.5, 167.8, 164.5, 135.7, 134.5, 133.3, 128.9, 125.5, 125.4, 122.8, 120.5, 118.7, 115.2, 110.7, 109.3, 51.7, 46.5, 37.0, 23.7, 13.3, 11.9, 10.4. . MS (ESI) m/z = 438.3 [M+H] +. HPLC (254 nm): tR = 1.823 min (95.7322%). (Z)-N-(2-(Diethylamino)ethyl)-5-((5-(ethylsulfonamido)-2-oxoindolin-3ylidene)methyl)-2,4-dimethyl-1H-pyrrole-3-carboxamide (4d). To a solution of (Z)-5-((5amino-2-oxoindolin-3-ylidene)methyl)-N-(2-(diethylamino)ethyl)-2,4-dimethyl-1H-


60

Page 61 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


pyrrole-3-carboxamide (100 mg, 0.25 mmol, 1.0 equiv) in dry DMF (2 ml) was added pyridine (20 µl, 0.25 mmol, 1 eq.) and ethane sulfonyl chloride (24 µl, 0.25 mmol, 1.0 equiv). The reaction mixture was stirred for 18 h at room temperature and then quenched with brine (20 ml) and ethyl acetate (30 ml) and washed with brine (5 x 30 ml). The organic layer was dried over Na2SO4 and the solvent was evaporated. The crude product was purified by reversed-phase flash chromatography (acetonitrile/H2O = 10 to 100%) to give the title compound as a yellow amorphous solid. Yield: 39 mg, 0. 08 mmol, 32 %. 1H NMR (500 MHz, CDCl3 / DMSO- d6 = 8/1) δH (ppm) = 13.37 (s, 1H), 9.83 (s, 1H), 8.72 (s, 1H), 7.58 (t, J = 5.7 Hz, 1H), 7.22 (s, 1H), 7.12 (s, 1H), 6.92 (d, J = 8.1 Hz, 1H), 6.68 (d, J = 8.2 Hz, 1H), 3.69 (q, J = 5.7 Hz, 2H), 3.24 (t, J = 5.7 Hz, 2H), 3.12 (q, J = 7.4 Hz, 4H), 2.93 (t, J = 7.3 Hz, 2H), 2.35 (s, 3H), 2.28 (s, 3H), 1.33 – 1.17 (m, 9H).

13C

NMR (126 MHz, CDCl3 / DMSO- d6 = 8/1) δC

(ppm) = 167.0, 141.4, 137.6, 136.0, 131.5, 129.8, 126.6, 126.2, 123.2, 120.9, 118.3, 115.4, 112.6, 110.0, 52.0, 48.0, 45.0, 35.3, 13.8, 10.9, 8.6, 8.0. MS (ESI) m/z = 488.24 [M+H] +. HPLC (254 nm): tR = 2.555 min (95.5186%). (Z)-N-(2-(2-(4-(2,8-Diethyl-5,5-difluoro-1,3,7,9-tetramethyl-5H-4λ4,5λ4-dipyrrolo[1,2c:2',1'-f][1,3,2]diazaborinin-10-yl)phenoxy)acetamido)ethyl)-2,4-dimethyl-5-((2-oxo-5(vinylsulfon-amido)indolin-3-ylidene)methyl)-1H-pyrrole-3-carboxamide (5a) To a solution of

(Z)-N-(2-(2-(4-(2,8-diethyl-5,5-difluoro-1,3,7,9-tetramethyl-5H-4l4,5l4-dipyrrolo[1,2-

c:2',1'-f][1,3,2]diazaborinin-10-yl)phenoxy)acetamido)ethyl)-2,4-dimethyl-5-((2-oxo-5(2,2,2-trifluoro-acetamido)indolin-3-ylidene)methyl)-1H-pyrrole-3-carboxamide

(56

mg,


61


Page 62 of 85

0.064 mmol, 1.0 equiv) in water (6 ml) and MeOH (8 ml) was added potassium carbonate (25 mg, 0.179 mmol, 2.8 equiv) and the reaction mixture was stirred at room temperature for 18 h. Then, saturated sodium bicarbonate solution (20 ml) and dichloromethane (20 ml) were added, and the aqueous layer was extracted with dichloromethane (2 x 20 ml). The combined organic layers were washed with brine (3 x 50 ml), dried over MgSO4 and filtered. The resulting solution was cooled to 0°C and 2-chloroethane sulfonyl chloride (10 µl, 0.1 mmol, 1.5 equiv) and triethylamine (18 µl, 0.13 mmol, 2.0 equiv) was added and the reaction mixture was stirred for 18 h at room temperature. The reaction was quenched with saturated aq. sodium bicarbonate solution (20 ml) and washed with brine (3 x 20 ml). The organic layer was dried over magnesium sulfate and the solvent was removed. The crude product was purified by silica chromatography (dichloromethane/MeOH = 20/1 to 10/1) to give the title compound as a red amorphous solid. Yield: 28 mg, 0.032 mmol, 51%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.68 (s, 1H), 10.89 (s, 1H), 9.54 (s, 1H), 8.31 (dd, J = 11.6, 6.1 Hz, 1H), 7.71 (d, J = 5.3 Hz, 1H), 7.55 (d, J = 3.6 Hz, 2H), 7.29 (d, J = 8.5 Hz, 2H), 7.19 (d, J = 8.5 Hz, 2H), 6.96 (dd, J = 8.2, 2.0 Hz, 1H), 6.84 (d, J = 8.2 Hz, 1H), 6.78 (dd, J = 16.4, 10.0 Hz, 1H), 6.06 – 5.96 (m, 2H), 4.59 (s, 2H), 3.37 (s, 5H), 2.46 (s, 3H), 2.44 (s, 7H), 2.43 (s, 3H), 2.31 (q, J = 7.5 Hz, 5H), 1.32 (s, 6H), 0.95 (t, J = 7.5 Hz, 6H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 169.5, 167.6, 165.1, 158.1, 153.0, 140.5, 138.1, 136.4, 136.1, 136.1, 136.0, 132.5, 130.9, 130.3, 129.8, 129.3, 127.4, 126.8, 126.0, 125.6, 123.7, 121.9, 120.5, 115.5, 114.7, 114.2, 109.6, 67.1, 38.7, 38.5, 16.4, 14.5, 13.3, 12.2, 11.5, 10.5. 11B NMR (160 MHz, DMSO-d6) δB (ppm) = 3.79


62

Page 63 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(t, J = 34.2 Hz) ppm.

19F

NMR (471 MHz, DMSO-d6) δF (ppm) = -142.99 (dd, J = 66.5, 29.5

Hz) ppm. MS (APCI) m/z = 846.36 [M-F] +. HPLC (360 nm): tR = 8.685 min (95.2191%). (Z)-5-((5-(2-Chloroacetamido)-2-oxoindolin-3-ylidene)methyl)-N-(2-(3-(5,5difluoro-7,9-dimethyl-5H-4λ4,5λ4-dipyrrolo[1,2-c:2',1'-f][1,3,2]diazaborinin-3yl)propanamido)ethyl)-2,4-dimethyl-1H-pyrrole-3-carboxamide (5b). To a solution of (Z)-N(2-(3-(5,5-difluoro-7,9-dimethyl-5H-4λ4,5λ4-dipyrrolo[1,2-c:2',1'-f][1,3,2]diazaborinin-3yl)propanamido)ethyl)-2,4-dimethyl-5-((2-oxo-5-(2,2,2-trifluoroacetamido)indolin-3ylidene)methyl)-1H-pyrrole-3-carboxamide (26 mg, 0.037 mmol, 1.0 equiv) in water (7 ml) and MeOH (15 ml) was added potassium carbonate (10.2 mg, 0.074 mmol, 2.0 equiv) and the reaction mixture was stirred at room temperature for 18 h. Then, saturated aq. sodium bicarbonate solution (30 ml) and dichloromethane (30 ml) were added, and the aqueous layer was extracted with dichloromethane (2 x 30 ml). The combined organic extracts were washed with brine (3 x 50 ml), dried over MgSO4 and filtered. The solution was cooled to 0°C and 2-chloro acetyl chloride (4.4 µl, 0.056 mmol, 1.5 equiv) and triethylamine (10.2 µl, 0.074 mmol, 2.0 equiv) was added and the reaction mixture was stirred for 18 h at room temperature. The reaction was quenched with saturated aq. sodium bicarbonate solution (20 ml) and washed with brine (3 x 20 ml). The organic layer was dried over magnesium sulfate and the solvent was removed. The crude product was purified by silica chromatography (dichloromethane/MeOH = 20/1) to give the title compound as a red amorphous solid. Yield: 7.6 mg, 0.011 mmol, 30%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.65 (s, 1H), 10.87 (s,


63


Page 64 of 85

1H), 10.14 (s, 1H), 8.06 (t, J = 5.5 Hz, 1H), 7.87 (d, J = 2.0 Hz, 1H), 7.66 (s, 1H), 7.55 (t, J = 5.4 Hz, 1H), 7.49 (s, 1H), 7.30 (dd, J = 8.3, 2.0 Hz, 1H), 7.06 (d, J = 4.0 Hz, 1H), 6.87 (d, J = 8.2 Hz, 1H), 6.37 (d, J = 4.0 Hz, 1H), 6.27 (s, 1H), 4.26 (s, 2H), 3.37 – 3.34 (m, 2H), 3.30 – 3.25 (m, 4H), 3.12 (t, J = 7.7 Hz, 2H), 2.48 (s, 3H), 2.46 (s, 3H), 2.41 (s, 3H), 2.26 (s, 3H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 171.2, 169.5, 164.9, 164.3, 159.1, 157.7, 144.0, 136.1, 135.1, 134.4, 133.0, 132.3, 129.4, 128.8, 125.6, 125.2, 123.2, 120.4, 120.2, 118.9, 116.5, 114.9, 111.0, 109.4, 43.5, 38.8, 38.5, 33.8, 24.0, 14.5, 13.3, 10.9, 10.5. MS (ESI) m/z = 712.3 [M+H] +. HPLC (360 nm): tR = 5.183 min (96.1730%). (Z)-N-(2-(2-(4-(2,8-Diethyl-5,5-difluoro-1,3,7,9-tetramethyl-5H-4λ4,5λ4dipyrrolo[1,2-c:2',1'-f][1,3,2]diazaborinin-10-yl)phenoxy)acetamido)ethyl)-2,4-dimethyl-5((2-oxo-5-(2,2,2-trifluoro-acetamido)indolin-3-ylidene)methyl)-1H-pyrrole-3-carboxamide (5c)

To

a

solution

of

2-(4-(2,8-diethyl-5,5-difluoro-1,3,7,9-tetramethyl-5H-4λ4,5λ4-

dipyrrolo[1,2-c:2',1'-f][1,3,2]-diazaborinin-10-yl)phenoxy)acetic acid (45 mg, 0.1 mmol, 1.0 equiv) in DMF was added 1-hydroxybenzotriazole hydrate (15 mg, 0.11 mmol, 1.1 equiv) and N-(3-dimethylaminopropyl)-N′-ethyl carbodiimide hydrochloride (20 mg, 0.1 mmol, 1.0 equiv) and the mixture was stirred for 30 min at room temperature. Then, a mixture of triethylamine (42 µl, 0.3 mmol, 3.0 equiv) and (Z)-N-(2-aminoethyl)-2,4-dimethyl-5-((2oxo-5-(2,2,2-trifluoroacetamido)-indolin-3-ylidene)methyl)-1H-pyrrole-3-carboxamide hydrochloride (47 mg, 0.1 mmol, 1.0 equiv) was added and the reaction mixture was stirred for 18 h at room temperature. The reaction mixture was diluted with ethyl acetate (30 ml)


64

Page 65 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and washed with 0.1 N hydrochloric acid, 0.1 N aq. sodium hydroxide solution and brine (each 3 x 30 ml). The combined organic extracts were dried over Na2SO4, the solvent was evaporated,

and

the

crude

product

was

purified

by

silica

chromatography

(dichloromethane/MeOH = 20/1) to give the title compound as a red amorphous solid. Yield: 56 mg, 0.064 mmol, 64%. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.65 (s, 1H), 11.12 (s, 1H), 10.96 (s, 1H), 8.30 (q, J = 7.5, 6.4 Hz, 1H), 7.95 (d, J = 2.0 Hz, 1H), 7.74 – 7.67 (m, 1H), 7.59 (s, 1H), 7.33 (dd, J = 8.3, 2.1 Hz, 1H), 7.32 – 7.24 (m, 2H), 7.22 – 7.15 (m, 2H), 6.93 (d, J = 8.3 Hz, 1H), 4.59 (s, 2H), 3.34 (s, 3H), 2.47 (s, 3H), 2.44 (d, J = 1.8 Hz, 8H), 2.30 (q, J = 7.4 Hz, 4H), 1.32 (s, 6H), 0.95 (t, J = 7.6 Hz, 6H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 169.6, 167.6, 165.1, 158.1, 154.6, 152.9, 140.5, 138.1, 136.3, 132.5, 130.3, 129.9, 129.7, 129.3, 127.4, 125.8, 125.7, 123.8, 120.7, 120.5, 117.2, 115.5, 114.9, 114.6, 113.0, 109.5, 67.1, 38.7, 38.5, 16.4, 14.5, 13.3, 12.2, 11.5, 10.4. 19F

11B

NMR (160 MHz, DMSO-d6) δB (ppm) = 3.69 (t, J = 33.6 Hz).

NMR (471 MHz, DMSO-d6) δF (ppm) = -73.89, -143.02 (dd, J = 65.7, 28.2 Hz) ppm. MS

(ESI) m/z = 852.37 [M-F] +. HPLC (360 nm): tR = 8.955 min (95.3825%). (Z)-5-((5-(Acetamido)-2-oxoindolin-3-ylidene)methyl)-N-(2-(3-(5,5-difluoro-7,9dimethyl-5H-4λ4,5λ4-dipyrrolo[1,2-c:2',1'-f][1,3,2]diazaborinin-3-yl)propanamido)ethyl)2,4-dimethyl-1H-pyrrole-3-carboxamide (5d). To a solution of (Z)-N-(2-(3-(5,5-difluoro-7,9dimethyl-5H-4λ4,5λ4-dipyrrolo[1,2-c:2',1'-f][1,3,2]diazaborinin-3-yl)propanamido)ethyl)2,4-dimethyl-5-((2-oxo-5-(2,2,2-trifluoroacetamido)indolin-3-ylidene)methyl)-1H-pyrrole3-carboxamide (15 mg, 0.021 mmol, 1.0 equiv) in water (5 ml) and MeOH (10 ml) was added


65


Page 66 of 85

potassium carbonate (5.8 mg, 0.042 mmol, 2.0 equiv) and the reaction mixture was stirred at room temperature for 18 h. Saturated aq. sodium bicarbonate solution (20 ml) and dichloromethane (20 ml) were added and the aqueous layer was extracted with dichloromethane (2 x 20 ml). The combined organic layers were washed with brine (3 x 50 ml), dried over MgSO4 and filtered. The solution was cooled to 0°C and acetyl chloride (2.2 µl, 0.0315 mmol, 1.5 equiv) and triethylamine (5.8 µl, 0.042 mmol, 2.0 equiv) were added and the reaction mixture was stirred for 18 h at room temperature. The reaction was quenched with saturated sodium bicarbonate solution (20 ml) and washed with brine (3 x 20 ml). The organic layer was dried over MgSO4 and the solvent was removed. The crude product was purified by silica chromatography (dichloromethane/MeOH = 20/1) to give the title compound as a red amorphous solid. Yield: 6.5 mg, 0.01 mmol, 48 %. 1H NMR (500 MHz, DMSO-d6) δH (ppm) = 13.62 (s, 1H), 10.79 (s, 1H), 9.75 (s, 1H), 8.03 (t, J = 5.4 Hz, 1H), 7.81 (d, J = 2.1 Hz, 1H), 7.65 (s, 1H), 7.52 (t, J = 5.3 Hz, 1H), 7.42 (s, 1H), 7.25 (dd, J = 8.3, 2.0 Hz, 1H), 7.05 (d, J = 4.1 Hz, 1H), 6.81 (d, J = 8.4 Hz, 1H), 6.35 (d, J = 4.1 Hz, 1H), 6.26 (s, 1H), 3.26 (t, J = 5.5 Hz, 2H), 3.10 (t, J = 7.7 Hz, 2H), 2.46 (s, 3H), 2.43 (s, 3H), 2.38 (s, 3H), 2.24 (s, 3H), 2.02 (s, 3H). 13C NMR (126 MHz, DMSO-d6) δC (ppm) = 171.16, 169.51, 167.79, 164.88, 159.11, 157.69, 144.00, 135.90, 134.47, 133.27, 132.96, 129.14, 128.81, 125.53, 125.39, 125.21, 122.83, 120.21, 118.68, 116.47, 115.21, 110.70, 109.29, 38.79, 38.51, 33.76, 28.98, 23.95, 23.73, 14.47, 13.30, 10.94, 10.43. MS (ESI) m/z = 678.3 [M+Na] +. HPLC (360 nm): tR = 5.597 min (99.8730%).


66

Page 67 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Biology Stock solutions of inhibitors 10 mM stock solutions of inhibitors in DMSO were made by dissolving 1 to 5 mg of solid in an appropriate amount of DMSO using an ultrasonic water bath (Compounds synthesized; Hypothemycin (Abcam); Sunitinib (LC Labs)) or were purchased as 10 mM DMSO Stock (Crenolanib, MedChemtronica). Cell lines Human Leukemia cell lines were purchased at DMSZ (MV4-11: ACC-102; THP-1 ACC-16; Jurkat: ACC-282) or provided by K. Stegmaier (NOMO1, U937, MOLM14). Cell lines were maintained in RPMI-1640 Medium supplemented with 10% FCS, 2 mM L-glutamine, penicillin (100 U/ml) and streptomycin (100µg/ml) at 37°C at 5% CO2. Cells were maintained at 0.25 to 1.5 x 106 Cells / ml. Cell viability assays Fluorescent Endpoint-Assay For determining the effect of inhibitors on leukemia cell lines, cells were harvested at densities between 0.5 to 1.5 x 106 Cells / ml and subsequently seeded with 25.000 cells / well in a 96 well-plate in 90 µl of RPMI-1640 medium with 0.1% FCS. After 16 h, cells were treated with vehicle or different concentrations of inhibitor (10 µM, 1 µM, 500 nM, 100 nM, 50 nM, 10 nM, 5 nM, 1 nM) in triplicates and incubated at 37°C at 5% CO2 for 72 h. Viable


67


Page 68 of 85

cells were quantified using the Celltiter Blue Assay (Promega) by adding 20 µl of Celltiter Blue reagent and incubation for 3 h at 37°C. Fluorescence was measured at a Tecan Infinite M1000 microplate reader (Exc. 530 nm, Em. 560 nm) and the relative inhibition of cell viability was calculated compared to vehicle control and cell-free control. Results were plotted to the logarithmic concentration of the inhibitor using Prism 7.02 (Graphpad San Diego, CA, USA) and the IC50 is determined using the equation “Y=100/(1+10((LogIC50X)*HillSlope))”.

Reported IC50 values are the mean values of two independent experiments

(Figures S48 – S52). Chemiluminescent Endpoint-Assay For determining the effect of fluorescent inhibitors on leukemia cell lines, cells were harvested at densities between 0.5 to 1.5 x 106 Cells / ml and subsequently seeded with 10.000 cells / well in a 384 well-plate in 20 µl of RPMI-1640 medium with 0.1% FCS. After 16 h, cells were treated with vehicle or various concentrations of inhibitor (10 µM, 7.5 µM, 5 µM, 2.5 µM, 1 µM, 500 nM, 100 nM) in triplicates and incubated at 37°C at 5% CO2 for 72 h. Viable Cells were quantified using the Celltiter-Glo 2.0 Assay (Promega) by adding 20 µl of Celltiter-Glo reagent to each well at room temperature and cell lysis for 2 min using the shaking function of the Tecan M1000 reader. Luminescence was measured after 10 minutes using 500 ms integration time and the relative inhibition of cell viability was calculated compared to vehicle control and cell-free control. Results were plotted to the logarithmic concentration of the inhibitor using Prism 7.02 (Graphpad San Diego, CA, USA) and IC50 is


68

Page 69 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


determined using the equation “Y=100/(1+10((LogIC50-X)*HillSlope))”. For determination of cytotoxicity in U937, MOLM14 and NOMO1 cells, cells were resuspended at 15,000 cells/mL and seeded at 40 µl/ well onto 384-well plates. Cells were then treated with a single agent and analyzed for cell viability on day 0 and day 3 post-treatment using the Cell-TiterGlo luminescent assay kit (Promega) according to the manufacturer’s protocol. Luminescence was read on a Fluostar Omega Reader (BMG Labtech). Chemiluminescent RealTime-Glo MT Cell Viability Assay For determining the time-dependent inhibition on FLT3-positive cells, MV4-11 cells were harvested at densities between 0.5 to 1.5 x 106 Cells / ml and subsequently seeded with 2.000 cells / well in a 384 well-plate in 27 µl of RPMI-1640 medium with 0.1% FCS. Cells were treated with vehicle or various concentrations of inhibitor (5 µM, 500 nM, 250 nM, 50 nM, 25 nM, 5 nM, 2.5 nM and 0.5 nM) in triplicates and incubated at 37°C at 5% CO2 for 77 h (Figure S39). Viable Cells were quantified using the RealTime-Glo MT Cell Viability Assay (Promega) by adding 30 µl of 2-fold reagent mixture to each well. Cell viability was determined at 2 h, 4 h, 6 h, 8 h, 10 h, 12 h, 14 h, 16 h, 18 h, 20 h, 25 h, 27 h, 29 h, 31 h, 33 h, 35 h, 37 h, 49 h, 54 h, 57 h and 77 h measuring the luminescence using 500 ms integration time. The relative inhibition of cell viability was calculated compared to vehicle control and cell-free control. Results were plotted to the logarithmic concentration of the inhibitor using Prism 7.02 (Graphpad San Diego, CA, USA) and IC50 is determined for each time-point indicated using the equation “Y=100/(1+10((LogIC50-X)*HillSlope))”.


69


Page 70 of 85

Western Blot Proteins were extracted using Lysis Buffer (Cell Signaling Technology) supplemented with Complete, EDTA-free Protease Inhibitor Cocktail (Roche Diagnostics). Protein samples were separated by SDS-PAGE and subsequently transferred to PVDF membranes, which were blocked in 5% BSA and incubated with primary antibodies against phospho FLT3 (Y969) (Cell Signaling Technology Cat. No. 3463), FLT3 (8F2) (Cell Signaling Technology Cat. No. 3462), phospho STAT5 (Y694) (Cell Signaling Technology Cat. No. 9351), STAT5 (Cell Signaling Technology Cat. No. 9363), GAPDH (Santa Cruz Biotechnology Cat. No. sc-47724). Membranes were washed in TBS-T and incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies.

Signal was detected by enhanced chemi-

luminescence (ThermoFisher Scientific). In-gel digest of FLT3 and MALDI-MS/MS analysis 2 µg of FLT3 wt diluted in 100 mM phosphate buffer (pH 7.4, total volume 100 µl) were incubated with 100 µM of compound for 5 h at 37°C. Samples were freeze-dried and denatured in 20 µl of 1x Lämmli buffer for 5 min at 95°C. Samples were purified by SDS gel electrophoresis (12% gel). Protein was stained using colloidal coomassie and the gel was destained using desalted water. Stained proteins in height of the 73 kDa marker were excised in 1 mm3 cubes and washed alternatingly in 100 mM ammoniumbicarbonate buffer and acetonitrile (200 µl, 4 times each). 100 µl of 10 mM dithiothreitol in ammoniumbicarbonate buffer was added and incubated for 25 min at 60°C. The supernatant was discarded, and the


70

Page 71 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


gel pieces were washed alternatingly in 100 mM ammoniumbicarbonate buffer and acetonitrile

(4

times

each).

100

µl

of

10

mM

iodoacetamide

solution

in

ammoniumbicarbonate buffer were added and incubated at 37°C in the dark for 40 min. The supernatant was discarded, and the gel pieces were washed alternatingly in 100 mM ammoniumbicarbonate buffer and acetonitrile (4 times each). The gel pieces were rehydrated in 20 µl of 20 ng/µl Trypsin/Lys C mix (Promega) for 10 min, then covered with 60 µl 25 mM ammoniumbicarbonate buffer and incubated overnight. 10 µl of 1% formic acid was added and the supernatant was lyophilized. Samples were reconstituted in 10 µl of 0.1% trifluoro acetic acid and purified using µ-C18 ZipTip tips (Merck Millipore). Samples were spotted in a α-cyano-4-hydroxycinnamic acid matrix and spectra were recorded on a Bruker Daltonics autoFlex MALDI TOF spectrometer using positive polarity. Spectra were analyzed using Mmass 5. Kinase inhibition assays Determination of kinase inhibition was performed using the ADP Glo Assay (Promega) or determined by external services (Eurofins or DiscoverX) as indicated. ADP Glo assay The ADP Glo assay was used as described by the supplier. Active kinase was purchased from Promega (FLT3-D835Y, 50 ng per assay), Invitrogen (FLT3-ITD, 100 ng per assay) and ProQinase (FLT3-wt, 100 ng per assay). Kinase was mixed with 10-fold inhibitor and preincubated for the time indicated at each assay (5 to 30 min). Kinase Reaction Buffer A (4 µl)


71


Page 72 of 85

and a mixture (4 µl) of myelobasic protein (MBP, final concentration 0.1 mg/ml) and ATP (final concentration 100 µM to 1 mM) was added and the kinase reaction was incubated at room temperature for 120 min. ADP Glo reagent (10 µl) was added and incubated for 40 min at room temperature. Then, the kinase detection reagent (20 µl) was added and incubated for 30 min (100 µM ATP) to 60 min (1 mM ATP) at room temperature. Relative Inhibition was calculated compared to vehicle control and reagent control and were plotted as relative inhibition of kinase activity to the logarithmic concentration of the inhibitor using Prism 7.02 (Graphpad, San Diego, CA, USA). IC50 is determined using the equation “Y=100/(1+10((LogIC50-X)*HillSlope))”. Ki and kinact of compounds 4a and 4b have been determined using the ADP Glo assay with pre-incubation times of 2 min, 10 min, 20 min and 30 min. The time-dependent IC50 values were fitted to the following equation using nonlinear regression: y=(((1+(S/KM))*KI)/(((kinact*(y/((KI*(1+(S/KM)))+y)))*x)/((2-((kinact*(y/((KI*(1+(S/KM)))+y)))*x))(2*exp((((-1)*kinact)*(y/((KI*(1+(S/KM)))+y)))*x))))) where KI and kinact can be calculated as reported by Krippendorf et al.42 using GraphPad Prism 8.0.1. Kinase Screening at Eurofins Pharma Drug Discovery Series and DiscoverX

Kd measurements for 4b were performed using the KINOMEscan® Assay Platform (DiscoverX, San Diego, United States). Assays were run using an 11-point 3-fold dilutional series starting at 30 µM inhibitor. Single-point determination of residual kinase activity for 4b was performed using the KinaseProfiler platform (Eurofins, Dundee, United Kingdom).


72

Page 73 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Assays were run using 10 µM ATP and kinase activity was measured by scintillation. Relative inhibition was calculated compared to vehicle control and reagent control. Kinase tree was illustrated using KinMap.36 Zebrafish Maintenance and Husbandry Adult zebrafish wildtype (Danio rerio) are maintained in 60 l fish tanks in a laboratory with no daylight and a constant temperature of 28°C. Room light is programmed to a 12 h dark / 12 h light cycle. Adult zebrafish are bred in mating containers (2 animals / 1 l) or mating tanks (6 – 8 animals / 20 l) that are equipped with plastic grass and a removable sieve at the bottom of the tank. Embryos are collected 2 to 3 hours after a light cycle started, rinsed with E3-Medium and isolated in petri dishes containing E3-Medium. Living embryos were sorted randomly 24 hours past fertilization in 96 well plates (n=1 / Well) or 24 Well plate (n = 10 / Well). Fish keeping protocols are approved by the Darmstadt administrative authority and documented. All fish were treated humanely. Danio rerio Toxicity Assay Stock solutions (10 mM in DMSO) of test compounds are diluted to 10x final concentrations. Embryos are sorted in 180 µl of E3 Medium and 24 hpf test compounds are added to give the final 1x concentrations. Survival was monitored every 24 h by observation of heart-beat. Embryos were euthanized 120 h past fertilization using eugenol. Survival rates were calculated compared to 0.1 % DMSO-treated controls and fish water controls. Images were recorded using a Panasonic Lumix GX8 equipped with a Lumix G macro 1:2.8/30 ASPH


73


Page 74 of 85

objective and a 10 mm and 16 mm extension tube (Meike MK-P-AF3B). Fluorescence was recorded using an AxioScopeA.1 microscope (Zeiss) equipped with an Axiocam MRc or equipped with a Nuance FX multispectral imaging system. Determination of compound half-life time Inhibitor half-life has been determined using 1.5 ml HPLC vials injecting 20 µl per run using the HPLC method as described above. All buffers are sonicated for 30 min in an open flask and have been purged with argon for 6 hours. 5 µl of a 1 mM DMSO Stock solution of inhibitor was mixed with 45 µl of acetonitrile, 50 µl of a 10 mM solution of nucleophile in 100 mM potassium phosphate buffer at the pH value indicated and 400 µl of 100 mM potassium phosphate buffer under argon immediately before the first injection into the HPLC. The samples have been measured at t = 0 min, 34 min, 68 min, 102 min, 136 min, 170 min, 204 min. The area of the signal corresponding to the inhibitor was determined (mAU·s) and the relative area compared to the first measurement was plotted to the corresponding time-point. The half-life was determined using one-phase exponential decay fit of Graphpad Prism 7.02 (Y = (Y0 - Plateau)·exp(-K*X) + Plateau). Half-life is calculated as t1/2 = ln (2) / k. Detection of fluorescence using SDS-PAGE and subsequent Western blotting 500 ng FLT3-D835Y (Promega, 10 µl) is diluted in 100 mM phosphate buffer (pH 7.4, total volume 100 µl) and is incubated with 100 µM of compound for 3 h at 37°C. Then 1 µl of 100fold DMSO stock of the fluorophore is added (final concentration 1 µM) and incubated for 2


74

Page 75 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


h at 37°C. Samples were freeze-dried and reconstituted with 10 µl of a 10 mM solution of GSH and incubated for 2 h at 37°C. Then the samples are denatured by adding 10 µl of 2x Lämmli buffer and incubation for 5 min at 95°C. Samples are subjected to SDS gel electrophoresis (12% gel) and subsequently blotted to a PVDF membrane using constant ampere (2 mA / cm2 gel, 45 min) and Western blot buffer (48 mM Tris, 39 mM glycine, 10% methanol). The blotting membrane was washed with water twice and dried in an air flow for 30 min. Fluorescence was recorded using a Vilber Fusion FX7 Edge imaging system using the settings indicated in the corresponding image.

ASSOCIATED CONTENT Supporting Information. Supporting Figures S1-S52 and Supporting Tables S1-S9. Synthesis and compound characterization data of synthesis intermediates, NMR spectra of test compounds, dose-response curves, molecular formula strings of compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author * [email protected] Author Contributions


75


Page 76 of 85

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding Sources This work was supported by the Leukemia & Lymphoma Society (K.St.) and the National Cancer Institute R35 CA210030-01 (K.St.). Notes K.St. participates in the DFCI/Novartis Drug Discovery Program which includes grant support for an unrelated project and previously included consulting and has consulted for Rigel Pharmaceuticals on a topic unrelated to this manuscript.

ACKNOWLEDGMENT The authors thank Dr. Tobias Meckel and Anke Imrich for technical assistance; Dr. Christoph Scholz for helpful discussions, and Merck Lab for access to the Vilber Imaging system.

Authors will release the atomic coordinates and experimental data upon article publication.


76

Page 77 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ABBREVIATIONS FLT3, FMS-like tyrosine kinase 3; ITD, internal tandem duplications; TKI, tyrosine kinase inhibitor; CRG, chemical reactive group; HA, α-halo acetamide; AC, acrylamide; VS, vinyl sulfonamide; PA, propargyl amide; HY, hypothemycin; SU, sunitinib. REFERENCES 1.

De Kouchkovsky, I.; Abdul-Hay, M., ‘Acute myeloid leukemia: a comprehensive

review and 2016 update’. Blood Cancer J 2016, 6, e441. 2.

Smith, C. C.; Wang, Q.; Chin, C. S.; Salerno, S.; Damon, L. E.; Levis, M. J.; Perl, A. E.;

Travers, K. J.; Wang, S.; Hunt, J. P.; Zarrinkar, P. P.; Schadt, E. E.; Kasarskis, A.; Kuriyan, J.; Shah, N. P., Validation of ITD mutations in FLT3 as a therapeutic target in human acute myeloid leukaemia. Nature 2012, 485 (7397), 260-263. 3.

Smith, C. C.; Lin, K.; Stecula, A.; Sali, A.; Shah, N. P., FLT3 D835 mutations confer

differential resistance to type II FLT3 inhibitors. Leukemia 2015, 29, 2390-2392. 4.

Stone, R. M.; Manley, P. W.; Larson, R. A.; Capdeville, R., Midostaurin: its odyssey

from discovery to approval for treating acute myeloid leukemia and advanced systemic mastocytosis. Blood Adv. 2018, 2 (4), 444-453. 5.

Barf, T.; Kaptein, A., Irreversible protein kinase inhibitors: balancing the benefits and

risks. J. Med. Chem. 2012, 55 (14), 6243-6262.


77


6.

Page 78 of 85

Dungo, R. T.; Keating, G. M., Afatinib: First global approval. Drugs 2013, 73 (13),

1503-1515. 7.

Neratinib approved for HER2+ Breast Cancer. Cancer discovery 2017, 7 (9), Of1.

8.

Greig, S. L., Osimertinib: first global approval. Drugs 2016, 76 (2), 263-273.

9.

de Claro, R. A.; McGinn, K. M.; Verdun, N.; Lee, S.-L.; Chiu, H.-J.; Saber, H.; Brower,

M. E.; Chang, C. J. G.; Pfuma, E.; Habtemariam, B.; Bullock, J.; Wang, Y.; Nie, L.; Chen, X.H.; Lu, D.; Al-Hakim, A.; Kane, R. C.; Kaminskas, E.; Justice, R.; Farrell, A. T.; Pazdur, R., FDA approval: Ibrutinib for patients with previously preated mantle cell lymphoma and previously treated chronic lymphocytic leukemia. Clin. Cancer Res. 2015, 21 (16), 35863590. 10. Schirmer, A.; Kennedy, J.; Murli, S.; Reid, R.; Santi, D. V., Targeted covalent inactivation of protein kinases by resorcylic acid lactone polyketides. Proc. of the Nat. Acad.

Sci. U. S. A. 2006, 103 (11), 4234-4239. 11. Chaikuad, A.; Koch, P.; Laufer, S. A.; Knapp, S., The cysteinome of protein kinases as a target in drug development. Angew. Chem. Int. Ed. 2018, 57 (16), 4372-4385. 12. Rastelli, G.; Rosenfeld, R.; Reid, R.; Santi, D. V., Molecular modeling and crystal structure of ERK2–hypothemycin complexes. J. Struct. Biol. 2008, 164 (1), 18-23.


78

Page 79 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


13. Wissner, A.; Floyd, M. B.; Johnson, B. D.; Fraser, H.; Ingalls, C.; Nittoli, T.; Dushin, R. G.; Discafani, C.; Nilakantan, R.; Marini, J.; Ravi, M.; Cheung, K.; Tan, X.; Musto, S.; Annable, T.; Siegel, M. M.; Loganzo, F., 2-(Quinazolin-4-ylamino)-[1,4]benzoquinones as covalent-binding, rrreversible rnhibitors of the kinase domain of vascular endothelial growth factor receptor-2. J. Med. Chem. 2005, 48 (24), 7560-7581. 14. Xu, J.; Ong, E. H. Q.; Hill, J.; Chen, A.; Chai, C. L. L., Design, synthesis and biological evaluation of FLT3 covalent inhibitors with a resorcylic acid core. Biorg. Med. Chem. 2014,

22 (23), 6625-6637. 15. Scarpino, A.; Ferenczy, G. G.; Keserű, G. M., Comparative evaluation of covalent docking tools. J. Chem. Inf. Model. 2018, 58 (7), 1441-1458. 16. Scholz, C.; Knorr, S.; Hamacher, K.; Schmidt, B., DOCKTITE—A highly versatile step-by-step workflow for covalent docking and virtual screening in the Molecular Operating Environment. J. Chem. Inf. Model. 2015, 55 (2), 398-406. 17. Molecular Operating Environment (MOE), 2013.08; Chemical Computing Group

ULC, 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7, 2018. 18. Latorre, A.; Schirmeister, T.; Kesselring, J.; Jung, S.; Johé, P.; Hellmich, U. A.; Heilos, A.; Engels, B.; Krauth-Siegel, R. L.; Dirdjaja, N.; Bou-Iserte, L.; Rodríguez, S.; González, F. V., Dipeptidyl nitroalkenes as potent reversible inhibitors of cysteine proteases rhodesain and cruzain. ACS Med. Chem. Lett. 2016, 7 (12), 1073-1076.


79


Page 80 of 85

19. Schirmeister, T.; Schmitz, J.; Jung, S.; Schmenger, T.; Krauth-Siegel, R. L.; Gütschow, M., Evaluation of dipeptide nitriles as inhibitors of rhodesain, a major cysteine protease of Trypanosoma brucei. Bioorg. Med. Chem. Lett. 2017, 27 (1), 45-50. 20. Royo, S.; Schirmeister, T.; Kaiser, M.; Jung, S.; Rodríguez, S.; Bautista, J. M.; González, F. V., Antiprotozoal and cysteine proteases inhibitory activity of dipeptidyl enoates. Biorg.

Med. Chem. 2018, 26 (16), 4624-4634. 21. LaMonte, G. M.; Almaliti, J.; Bibo-Verdugo, B.; Keller, L.; Zou, B. Y.; Yang, J.; Antonova-Koch, Y.; Orjuela-Sanchez, P.; Boyle, C. A.; Vigil, E.; Wang, L.; Goldgof, G. M.; Gerwick, L.; O’Donoghue, A. J.; Winzeler, E. A.; Gerwick, W. H.; Ottilie, S., Development of a potent inhibitor of the plasmodium proteasome with reduced mammalian toxicity. J. Med.

Chem. 2017, 60 (15), 6721-6732. 22. London, N.; Miller, R. M.; Krishnan, S.; Uchida, K.; Irwin, J. J.; Eidam, O.; Gibold, L.; Cimermančič, P.; Bonnet, R.; Shoichet, B. K.; Taunton, J., Covalent docking of large libraries for the discovery of chemical probes. Nature Chem. Biol. 2014, 10, 1066-1072. 23. Shraga, A.; Olshvang, E.; Davidzohn, N.; Khoshkenar, P.; Germain, N.; Shurrush, K.; Carvalho, S.; Avram, L.; Albeck, S.; Unger, T.; Lefker, B.; Subramanyam, C.; Hudkins, R. L.; Mitchell, A.; Shulman, Z.; Kinoshita, T.; London, N., Covalent docking identifies a potent and selective MKK7 Inhibitor. Cell chem biol 2019, 26 (1), 98-108.e5.


80

Page 81 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


24. Mol, C. D.; Lim, K. B.; Sridhar, V.; Zou, H.; Chien, E. Y. T.; Sang, B.-C.; Nowakowski, J.; Kassel, D. B.; Cronin, C. N.; McRee, D. E., Structure of a c-Kit product complex reveals the basis for kinase transactivation. J. Biol. Chem. 2003, 278 (34), 31461-31464. 25. Larrosa-Garcia, M.; Baer, M. R., FLT3 Inhibitors in acute myeloid leukemia: current status and future directions. Mol. Cancer Ther. 2017, 16 (6), 991-1001. 26. Liu, T.; Lin, Y.; Wen, X.; Jorissen, R. N.; Gilson, M. K., BindingDB: a web-accessible database of experimentally determined protein–ligand binding affinities. Nucleic Acids Res. 2007, 35 (suppl_1), D198-D201. 27. Liao, J. J.-L., Molecular recognition of protein kinase binding pockets for design of potent and selective kinase inhibitors. J. Med. Chem. 2007, 50 (3), 409-424. 28. Neudert, G.; Klebe, G., DSX: A knowledge-based scoring function for the assessment of protein–ligand complexes. J. of Chem. Inf. Model. 2011, 51 (10), 2731-2745. 29. Statsuk, A. V.; Maly, D. J.; Seeliger, M. A.; Fabian, M. A.; Biggs, W. H.; Lockhart, D. J.; Zarrinkar, P. P.; Kuriyan, J.; Shokat, K. M., Tuning a three-component reaction for trapping kinase substrate complexes. J. Am. Chem. Soc. 2008, 130 (51), 17568-17574. 30. Zhao, Q.; Ouyang, X.; Wan, X.; Gajiwala, K. S.; Kath, J. C.; Jones, L. H.; Burlingame, A. L.; Taunton, J., Broad-spectrum kinase profiling in live cells with lysine-targeted sulfonyl fluoride probes. J. Am. Chem. Soc. 2017, 139 (2), 680-685.


81


Page 82 of 85

31. Braunbeck, T.; Kais, B.; Lammer, E.; Otte, J.; Schneider, K.; Stengel, D.; Strecker, R., The fish embryo test (FET): origin, applications, and future. Environ. Sci. Pollut. Res. 2015,

22 (21), 16247-16261. 32. Fabian, M. A.; Biggs, W. H., 3rd; Treiber, D. K.; Atteridge, C. E.; Azimioara, M. D.; Benedetti, M. G.; Carter, T. A.; Ciceri, P.; Edeen, P. T.; Floyd, M.; Ford, J. M.; Galvin, M.; Gerlach, J. L.; Grotzfeld, R. M.; Herrgard, S.; Insko, D. E.; Insko, M. A.; Lai, A. G.; Lelias, J. M.; Mehta, S. A.; Milanov, Z. V.; Velasco, A. M.; Wodicka, L. M.; Patel, H. K.; Zarrinkar, P. P.; Lockhart, D. J., A small molecule-kinase interaction map for clinical kinase inhibitors.

Nat. Biotechnol. 2005, 23 (3), 329-336. 33. Mologni, L.; Rostagno, R.; Brussolo, S.; Knowles, P. P.; Kjaer, S.; Murray-Rust, J.; Rosso, E.; Zambon, A.; Scapozza, L.; McDonald, N. Q.; Lucchini, V.; Gambacorti-Passerini, C., Synthesis, structure–activity relationship and crystallographic studies of 3-substituted indolin-2-one RET inhibitors. Biorg. Med. Chem. 2010, 18 (4), 1482-1496. 34. Zimmerman, E. I.; Turner, D. C.; Buaboonnam, J.; Hu, S.; Orwick, S.; Roberts, M. S.; Janke, L. J.; Ramachandran, A.; Stewart, C. F.; Inaba, H.; Baker, S. D., Crenolanib is active against models of drug-resistant FLT3-ITD−positive acute myeloid leukemia. Blood 2013, 122 (22), 3607-3615.


82

Page 83 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


35. Davis, M. I.; Hunt, J. P.; Herrgard, S.; Ciceri, P.; Wodicka, L. M.; Pallares, G.; Hocker, M.; Treiber, D. K.; Zarrinkar, P. P., Comprehensive analysis of kinase inhibitor selectivity.

Nat. Biotechnol. 2011, 29, 1046-1051. 36. Eid, S.; Turk, S.; Volkamer, A.; Rippmann, F.; Fulle, S., KinMap: a web-based tool for interactive navigation through human kinome data. BMC Bioinformatics 2017, 18 (1), 16. 37. Wodicka, L. M.; Ciceri, P.; Davis, M. I.; Hunt, J. P.; Floyd, M.; Salerno, S.; Hua, X. H.; Ford, J. M.; Armstrong, R. C.; Zarrinkar, P. P.; Treiber, D. K., Activation state-dependent binding of small molecule kinase inhibitors: structural insights from biochemistry. Chem.

Biol. 2010, 17 (11), 1241-1249. 38. Turetsky, A.; Kim, E.; Kohler, R. H.; Miller, M. A.; Weissleder, R., Single cell imaging of brutons tyrosine kinase using an irreversible inhibitor. Sci. Rep. 2014, 4, 4782. 39. Sturgeon, J. B.; Laird, B. B., Symplectic algorithm for constant-pressure molecular dynamics using a Nosé–Poincaré thermostat. J. Chem. Phys. 2000, 112 (8), 3474-3482. 40. Amombo, G. M.; Kramer, T.; Lo Monte, F.; Goring, S.; Fach, M.; Smith, S.; Kolb, S.; Schubenel, R.; Baumann, K.; Schmidt, B., Modification of a promiscuous inhibitor shifts the inhibition from gamma-secretase to FLT-3. Bioorg Med Chem Lett 2012, 22 (24), 7634-7640. 41. Fu, H. Y.; Chen, L.; Doucet, H., Phosphine-free palladium-catalyzed direct arylation of imidazo[1,2-a]pyridines with aryl bromides at low catalyst loading. J. Org. Chem. 2012, 77 (9), 4473-4478.


83


Page 84 of 85

42. Krippendorff, B. F.; Neuhaus, R.; Lienau, P.; Reichel, A.; Huisinga, W., Mechanismbased inhibition: deriving K(I) and k(inact) directly from time-dependent IC(50) values. J.

biomol. screen. 2009, 14 (8), 913-923.


84

Page 85 of 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents graphic


85

Virtual Screening Identifies Irreversible FMS-like ... - ACS Publications

Recommend Documents